deepbox 0.1.0

package/dist/chunk-XMWVME2W.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/stats/index.ts","../src/stats/_internal.ts","../src/stats/correlation.ts","../src/stats/descriptive.ts","../src/stats/tests.ts"],"names":["outShape","mean","out","cov","z","variance","std","quantile","F"],"mappings":";;;;AAAA,IAAA,aAAA,GAAA;AAAA,QAAA,CAAA,aAAA,EAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,SAAA,EAAA,MAAA,SAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,GAAA,EAAA,MAAA,GAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,iBAAA,EAAA,MAAA,iBAAA;AAAA,EAAA,aAAA,EAAA,MAAA,aAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,UAAA,EAAA,MAAA,UAAA;AAAA,EAAA,OAAA,EAAA,MAAA,OAAA;AAAA,EAAA,MAAA,EAAA,MAAA,MAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,MAAA,EAAA,MAAA,MAAA;AAAA,EAAA,YAAA,EAAA,MAAA,YAAA;AAAA,EAAA,IAAA,EAAA,MAAA,IAAA;AAAA,EAAA,MAAA,EAAA,MAAA,MAAA;AAAA,EAAA,IAAA,EAAA,MAAA,IAAA;AAAA,EAAA,MAAA,EAAA,MAAA,MAAA;AAAA,EAAA,UAAA,EAAA,MAAA,UAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,UAAA,EAAA,MAAA,UAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,OAAA,EAAA,MAAA,OAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,SAAA,EAAA,MAAA,SAAA;AAAA,EAAA,GAAA,EAAA,MAAA,GAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,WAAA,EAAA,MAAA,WAAA;AAAA,EAAA,SAAA,EAAA,MAAA,SAAA;AAAA,EAAA,SAAA,EAAA,MAAA,SAAA;AAAA,EAAA,QAAA,EAAA,MAAA,QAAA;AAAA,EAAA,QAAA,EAAA,MAAA;AAAA,CAAA,CAAA;;;AC0DO,SAAS,aAAA,CAAc,MAA4B,IAAA,EAAiC;AACzF,EAAA,IAAI,IAAA,KAAS,MAAA,EAAW,OAAO,EAAC;AAChC,EAAA,MAAM,SAAA,GAAoB,KAAA,CAAM,OAAA,CAAQ,IAAI,CAAA,GAAI,CAAC,GAAG,IAAI,CAAA,GAAI,CAAC,IAAI,CAAA;AAEjE,EAAA,MAAM,IAAA,uBAAW,GAAA,EAAY;AAC7B,EAAA,MAAM,SAAmB,EAAC;AAE1B,EAAA,KAAA,MAAW,MAAM,SAAA,EAAW;AAC1B,IAAA,MAAM,IAAA,GAAO,aAAA,CAAc,EAAA,EAAI,IAAI,CAAA;AACnC,IAAA,IAAI,CAAC,IAAA,CAAK,GAAA,CAAI,IAAI,CAAA,EAAG;AACnB,MAAA,IAAA,CAAK,IAAI,IAAI,CAAA;AACb,MAAA,MAAA,CAAO,KAAK,IAAI,CAAA;AAAA,IAClB;AAAA,EACF;AAGA,EAAA,OAAO,OAAO,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACpC;AAsBO,SAAS,YAAA,CAAa,KAAA,EAAc,IAAA,EAAyB,QAAA,EAA0B;AAC5F,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,OAAO,QAAA,GAAW,IAAI,KAAA,CAAc,KAAA,CAAM,MAAM,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,EAAC;AAAA,EAC/D;AAEA,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,MAAgB,EAAC;AAEvB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,IAAA,MAAM,CAAA,GAAI,MAAM,CAAC,CAAA;AACjB,IAAA,IAAI,CAAA,KAAM,MAAA,EAAW,MAAM,IAAI,WAAW,yCAAyC,CAAA;AAEnF,IAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACjB,MAAA,IAAI,QAAA,EAAU,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA;AAAA,IAC1B,CAAA,MAAO;AACL,MAAA,GAAA,CAAI,KAAK,CAAC,CAAA;AAAA,IACZ;AAAA,EACF;AAEA,EAAA,aAAA,CAAc,GAAG,CAAA;AACjB,EAAA,OAAO,GAAA;AACT;AAkBO,SAAS,eAAe,KAAA,EAA6C;AAC1E,EAAA,MAAM,OAAA,GAAU,IAAI,KAAA,CAAc,KAAA,CAAM,MAAM,CAAA;AAC9C,EAAA,IAAI,MAAA,GAAS,CAAA;AACb,EAAA,KAAA,IAAS,IAAI,KAAA,CAAM,MAAA,GAAS,CAAA,EAAG,CAAA,IAAK,GAAG,CAAA,EAAA,EAAK;AAC1C,IAAA,OAAA,CAAQ,CAAC,CAAA,GAAI,MAAA;AACb,IAAA,MAAA,IAAU,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAAA,EACxB;AACA,EAAA,OAAO,OAAA;AACT;AAmBO,SAAS,cAAA,CAAe,CAAA,EAAW,CAAA,EAAW,IAAA,EAAoB;AACvE,EAAA,IAAI,CAAA,CAAE,IAAA,KAAS,CAAA,CAAE,IAAA,EAAM;AACrB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,GAAG,IAAI,CAAA,qDAAA,EAAwD,EAAE,IAAI,CAAA,KAAA,EAAQ,EAAE,IAAI,CAAA,CAAA;AAAA,MACnF,MAAA;AAAA,MACA,EAAE,CAAA,EAAG,CAAA,CAAE,IAAA,EAAM,CAAA,EAAG,EAAE,IAAA;AAAK,KACzB;AAAA,EACF;AACF;AAmBO,SAAS,WAAA,CAAY,GAAW,MAAA,EAAwB;AAC7D,EAAA,IAAI,EAAE,KAAA,KAAU,QAAA,IAAY,MAAM,OAAA,CAAQ,CAAA,CAAE,IAAI,CAAA,EAAG;AACjD,IAAA,MAAM,IAAI,WAAW,0CAA0C,CAAA;AAAA,EACjE;AAEA,EAAA,OAAO,kBAAA,CAAmB,CAAA,CAAE,IAAA,EAAM,MAAM,CAAA;AAC1C;AAWO,SAAS,SAAS,MAAA,EAGvB;AACA,EAAA,MAAM,IAAI,MAAA,CAAO,MAAA;AACjB,EAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,CAAC,CAAA;AAChC,EAAA,IAAI,MAAM,CAAA,EAAG,OAAO,EAAE,KAAA,EAAO,QAAQ,CAAA,EAAE;AAEvC,EAAA,MAAM,SAAS,KAAA,CAAM,IAAA,CAAK,QAAQ,CAAC,CAAA,EAAG,OAAO,EAAE,CAAA,EAAG,GAAE,CAAE,CAAA,CAAE,KAAK,CAAC,CAAA,EAAG,MAAM,CAAA,CAAE,CAAA,GAAI,EAAE,CAAC,CAAA;AAChF,EAAA,IAAI,MAAA,GAAS,CAAA;AAEb,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,MAAA,IAAU;AACnC,IAAA,IAAI,IAAI,CAAA,GAAI,CAAA;AACZ,IAAA,OAAO,CAAA,GAAI,MAAA,CAAO,MAAA,IAAU,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,MAAA,CAAO,CAAC,CAAA,EAAG,CAAA,EAAG;AACzD,MAAA,CAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,IAAI,CAAA,GAAI,CAAA;AACd,IAAA,MAAM,OAAA,GAAA,CAAW,CAAA,GAAI,CAAA,GAAI,CAAA,IAAK,CAAA;AAC9B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,GAAA,GAAM,MAAA,CAAO,CAAC,CAAA,EAAG,CAAA;AACvB,MAAA,IAAI,QAAQ,MAAA,EAAW;AACrB,QAAA,KAAA,CAAM,GAAG,CAAA,GAAI,OAAA;AAAA,MACf;AAAA,IACF;AACA,IAAA,IAAI,CAAA,GAAI,CAAA,EAAG,MAAA,IAAU,CAAA,GAAI,IAAI,CAAA,GAAI,CAAA;AACjC,IAAA,CAAA,GAAI,CAAA;AAAA,EACN;AAEA,EAAA,OAAO,EAAE,OAAO,MAAA,EAAO;AACzB;AAuBO,SAAS,kBAAA,CACd,GACA,EAAA,EACM;AACN,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAElB,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,EAAA,CAAG,CAAA,CAAE,MAAA,EAAQ,EAAE,CAAA;AACf,IAAA;AAAA,EACF;AAEA,EAAA,MAAM,QAAQ,CAAA,CAAE,KAAA;AAChB,EAAA,MAAM,UAAU,CAAA,CAAE,OAAA;AAClB,EAAA,MAAM,MAAM,IAAI,KAAA,CAAc,EAAE,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA;AAC5C,EAAA,IAAI,SAAS,CAAA,CAAE,MAAA;AAEf,EAAA,OAAO,IAAA,EAAM;AACX,IAAA,EAAA,CAAG,QAAQ,GAAG,CAAA;AAGd,IAAA,IAAI,IAAA,GAAO,EAAE,IAAA,GAAO,CAAA;AACpB,IAAA,WAAS;AACP,MAAA,GAAA,CAAI,IAAI,CAAA,GAAA,CAAK,GAAA,CAAI,IAAI,KAAK,CAAA,IAAK,CAAA;AAC/B,MAAA,MAAA,IAAU,OAAA,CAAQ,IAAI,CAAA,IAAK,CAAA;AAE3B,MAAA,MAAM,GAAA,GAAM,KAAA,CAAM,IAAI,CAAA,IAAK,CAAA;AAC3B,MAAA,IAAA,CAAK,GAAA,CAAI,IAAI,CAAA,IAAK,CAAA,IAAK,GAAA,EAAK;AAG5B,MAAA,MAAA,IAAA,CAAW,IAAI,IAAI,CAAA,IAAK,CAAA,KAAM,OAAA,CAAQ,IAAI,CAAA,IAAK,CAAA,CAAA;AAC/C,MAAA,GAAA,CAAI,IAAI,CAAA,GAAI,CAAA;AACZ,MAAA,IAAA,EAAA;AACA,MAAA,IAAI,OAAO,CAAA,EAAG;AAAA,IAChB;AAAA,EACF;AACF;AAqBO,SAAS,UAAA,CAAW,CAAA,EAAW,IAAA,EAA4B,QAAA,EAA2B;AAC3F,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AAErB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,IACxF;AAEA,IAAA,IAAI,GAAA,GAAM,CAAA;AACV,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,GAAA,IAAO,WAAA,CAAY,GAAG,GAAG,CAAA;AAAA,IAC3B,CAAC,CAAA;AAED,IAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,CAAC,CAAA;AAC9B,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA,GAAM,CAAA,CAAE,IAAA;AAEjB,IAAA,MAAMA,SAAAA,GAAW,QAAA,GAAW,IAAI,KAAA,CAAc,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,EAAC;AACjE,IAAA,OAAO,OAAO,cAAA,CAAe;AAAA,MAC3B,IAAA,EAAM,GAAA;AAAA,MACN,KAAA,EAAOA,SAAAA;AAAA,MACP,KAAA,EAAO,SAAA;AAAA,MACP,QAAQ,CAAA,CAAE;AAAA,KACX,CAAA;AAAA,EACH;AAEA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,QAAQ,CAAA;AACrD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,OAAO,CAAA;AAErC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAY,IAAI,CAAA;AACnC,EAAA,MAAM,WAAA,GAAc,IAAA,CAAK,MAAA,CAAO,CAAC,GAAA,EAAK,EAAA,KAAO,GAAA,IAAO,CAAA,CAAE,KAAA,CAAM,EAAE,CAAA,IAAK,CAAA,CAAA,EAAI,CAAC,CAAA;AACxE,EAAA,IAAI,gBAAgB,CAAA,EAAG;AACrB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,+CAAA;AAAA,MACA,aAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,QAAA,EAAU;AAEZ,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,MAAM,CAAA,GAAI,UAAA,CAAW,CAAC,CAAA,IAAK,CAAA;AAC3B,QAAA,MAAM,CAAA,GAAI,OAAO,GAAA,CAAI,CAAC,IAAI,CAAA,GAAK,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACzC,QAAA,OAAA,IAAW,CAAA,GAAI,CAAA;AAAA,MACjB;AAAA,IACF,CAAA,MAAO;AACL,MAAA,IAAI,EAAA,GAAK,CAAA;AACT,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,QAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,QAAA,EAAA,EAAA;AAAA,MACF;AAAA,IACF;AAEA,IAAA,IAAA,CAAK,OAAO,KAAK,IAAA,CAAK,OAAO,KAAK,CAAA,IAAK,WAAA,CAAY,GAAG,GAAG,CAAA;AAAA,EAC3D,CAAC,CAAA;AAED,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,IAAA,IAAA,CAAK,CAAC,CAAA,GAAA,CAAK,IAAA,CAAK,CAAC,KAAK,CAAA,IAAK,WAAA;AAAA,EAC7B;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,IAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAuBO,SAAS,cAAA,CACd,CAAA,EACA,IAAA,EACA,QAAA,EACA,IAAA,EACQ;AACR,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AAEvC,EAAA,IAAI,CAAA,CAAE,UAAU,QAAA,EAAU;AACxB,IAAA,MAAM,IAAI,WAAW,2CAA2C,CAAA;AAAA,EAClE;AAEA,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,IAC5F;AACA,IAAA,IAAI,OAAO,CAAA,EAAG;AACZ,MAAA,MAAM,IAAI,qBAAA,CAAsB,2BAAA,EAA6B,MAAA,EAAQ,IAAI,CAAA;AAAA,IAC3E;AACA,IAAA,IAAI,CAAA,CAAE,QAAQ,IAAA,EAAM;AAClB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,CAAA,KAAA,EAAQ,IAAI,CAAA,SAAA,EAAY,CAAA,CAAE,IAAI,CAAA,oBAAA,CAAA;AAAA,QAC9B,MAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAIA,IAAA,IAAIC,KAAAA,GAAO,CAAA;AACX,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,IAAI,CAAA,GAAI,CAAA;AACR,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,MAAA,CAAA,EAAA;AACA,MAAA,MAAM,QAAQ,CAAA,GAAIA,KAAAA;AAClB,MAAAA,SAAQ,KAAA,GAAQ,CAAA;AAChB,MAAA,MAAM,SAAS,CAAA,GAAIA,KAAAA;AACnB,MAAA,EAAA,IAAM,KAAA,GAAQ,MAAA;AAAA,IAChB,CAAC,CAAA;AAED,IAAA,MAAMC,IAAAA,GAAM,IAAI,YAAA,CAAa,CAAC,CAAA;AAC9B,IAAAA,IAAAA,CAAI,CAAC,CAAA,GAAI,EAAA,IAAM,CAAA,GAAI,IAAA,CAAA;AAEnB,IAAA,MAAMF,SAAAA,GAAW,QAAA,GAAW,IAAI,KAAA,CAAc,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,EAAC;AACjE,IAAA,OAAO,OAAO,cAAA,CAAe;AAAA,MAC3B,IAAA,EAAME,IAAAA;AAAA,MACN,KAAA,EAAOF,SAAAA;AAAA,MACP,KAAA,EAAO,SAAA;AAAA,MACP,QAAQ,CAAA,CAAE;AAAA,KACX,CAAA;AAAA,EACH;AAEA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,QAAQ,CAAA;AACrD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAElD,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAY,IAAI,CAAA;AACnC,EAAA,MAAM,WAAA,GAAc,IAAA,CAAK,MAAA,CAAO,CAAC,GAAA,EAAK,EAAA,KAAO,GAAA,IAAO,CAAA,CAAE,KAAA,CAAM,EAAE,CAAA,IAAK,CAAA,CAAA,EAAI,CAAC,CAAA;AACxE,EAAA,IAAI,gBAAgB,CAAA,EAAG;AACrB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,mDAAA;AAAA,MACA,aAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,IAAI,OAAO,CAAA,EAAG;AACZ,IAAA,MAAM,IAAI,qBAAA,CAAsB,2BAAA,EAA6B,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC3E;AACA,EAAA,IAAI,eAAe,IAAA,EAAM;AACvB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,CAAA,KAAA,EAAQ,IAAI,CAAA,iBAAA,EAAoB,WAAW,CAAA,oBAAA,CAAA;AAAA,MAC3C,MAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,OAAO,CAAA;AACtC,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,QAAA,EAAU;AACZ,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,MAAM,CAAA,GAAI,UAAA,CAAW,CAAC,CAAA,IAAK,CAAA;AAC3B,QAAA,MAAM,CAAA,GAAI,OAAO,GAAA,CAAI,CAAC,IAAI,CAAA,GAAK,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACzC,QAAA,OAAA,IAAW,CAAA,GAAI,CAAA;AAAA,MACjB;AAAA,IACF,CAAA,MAAO;AACL,MAAA,IAAI,EAAA,GAAK,CAAA;AACT,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,QAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,QAAA,EAAA,EAAA;AAAA,MACF;AAAA,IACF;AAEA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,MAAM,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,CAAA,IAAK,CAAA,IAAK,CAAA;AACnC,IAAA,MAAA,CAAO,OAAO,CAAA,GAAI,CAAA;AAElB,IAAA,MAAMC,KAAAA,GAAO,KAAA,CAAM,OAAO,CAAA,IAAK,CAAA;AAC/B,IAAA,MAAM,QAAQ,CAAA,GAAIA,KAAAA;AAClB,IAAA,MAAM,QAAA,GAAWA,QAAO,KAAA,GAAQ,CAAA;AAChC,IAAA,KAAA,CAAM,OAAO,CAAA,GAAI,QAAA;AACjB,IAAA,MAAM,SAAS,CAAA,GAAI,QAAA;AACnB,IAAA,GAAA,CAAI,OAAO,CAAA,GAAA,CAAK,GAAA,CAAI,OAAO,CAAA,IAAK,KAAK,KAAA,GAAQ,MAAA;AAAA,EAC/C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,GAAA,CAAI,CAAC,CAAA,GAAA,CAAK,GAAA,CAAI,CAAC,CAAA,IAAK,MAAM,CAAA,GAAI,IAAA,CAAA;AAAA,EAChC;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAQA,IAAM,cAAA,GAAoC;AAAA,EACxC,kBAAA;AAAA,EAAqB,iBAAA;AAAA,EAAmB,mBAAA;AAAA,EAAqB,iBAAA;AAAA,EAC7D,kBAAA;AAAA,EAAoB,kBAAA;AAAA,EAAoB,oBAAA;AAAA,EAAsB,oBAAA;AAAA,EAC9D;AACF,CAAA;AAkBO,SAAS,SAAS,CAAA,EAAmB;AAC1C,EAAA,IAAI,IAAI,GAAA,EAAK;AAGX,IAAA,OAAO,KAAK,GAAA,CAAI,IAAA,CAAK,EAAE,CAAA,GAAI,KAAK,GAAA,CAAI,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,KAAK,CAAC,CAAC,CAAA,GAAI,QAAA,CAAS,IAAI,CAAC,CAAA;AAAA,EAC7E;AAGA,EAAA,CAAA,IAAK,CAAA;AACL,EAAA,IAAI,CAAA,GAAI,cAAA,CAAe,CAAC,CAAK;AAE7B,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,cAAA,CAAe,QAAQ,CAAA,EAAA,EAAK;AAC9C,IAAA,CAAA,IAAA,CAAM,cAAA,CAAe,CAAC,CAAA,IAAK,CAAA,KAAM,CAAA,GAAI,CAAA,CAAA;AAAA,EACvC;AAEA,EAAA,MAAM,IAAI,CAAA,GAAI,GAAA;AAEd,EAAA,OAAO,MAAM,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,IAAA,CAAK,EAAE,CAAA,GAAA,CAAK,CAAA,GAAI,GAAA,IAAO,IAAA,CAAK,IAAI,CAAC,CAAA,GAAI,CAAA,GAAI,IAAA,CAAK,IAAI,CAAC,CAAA;AAC/E;AAaA,SAAS,MAAA,CAAO,CAAA,EAAW,CAAA,EAAW,CAAA,EAAmB;AACvD,EAAA,MAAM,QAAA,GAAW,GAAA;AACjB,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,MAAM,KAAA,GAAQ,MAAA;AAGd,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAGhB,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,IAAI,CAAA,GAAI,CAAA,GAAK,GAAA,GAAM,CAAA,GAAK,GAAA;AACxB,EAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,EAAA,CAAA,GAAI,CAAA,GAAI,CAAA;AACR,EAAA,IAAI,CAAA,GAAI,CAAA;AAGR,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,IAAK,QAAA,EAAU,CAAA,EAAA,EAAK;AAClC,IAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AAGf,IAAA,IAAI,KAAM,CAAA,IAAK,CAAA,GAAI,KAAK,CAAA,IAAA,CAAO,GAAA,GAAM,OAAO,CAAA,GAAI,EAAA,CAAA,CAAA;AAChD,IAAA,CAAA,GAAI,IAAI,EAAA,GAAK,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,IAAI,EAAA,GAAK,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,CAAA,GAAI,CAAA;AACR,IAAA,CAAA,IAAK,CAAA,GAAI,CAAA;AAGT,IAAA,EAAA,GAAM,EAAE,IAAI,CAAA,CAAA,IAAM,GAAA,GAAM,KAAK,CAAA,IAAA,CAAO,CAAA,GAAI,OAAO,GAAA,GAAM,EAAA,CAAA,CAAA;AACrD,IAAA,CAAA,GAAI,IAAI,EAAA,GAAK,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,IAAI,EAAA,GAAK,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,CAAA,GAAI,CAAA;AACR,IAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,IAAA,CAAA,IAAK,GAAA;AAGL,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,GAAA,GAAM,CAAG,IAAI,GAAA,EAAK;AAAA,EACjC;AAEA,EAAA,OAAO,CAAA;AACT;AAsBO,SAAS,yBAAA,CAA0B,CAAA,EAAW,CAAA,EAAW,CAAA,EAAmB;AACjF,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,KAAK,CAAA,EAAG;AACjC,IAAA,MAAM,IAAI,qBAAA,CAAsB,eAAA,EAAiB,GAAA,EAAK,CAAC,CAAA;AAAA,EACzD;AACA,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,KAAK,CAAA,EAAG;AACjC,IAAA,MAAM,IAAI,qBAAA,CAAsB,eAAA,EAAiB,GAAA,EAAK,CAAC,CAAA;AAAA,EACzD;AAEA,EAAA,IAAI,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,CAAA,EAAG;AAClB,IAAA,MAAM,IAAI,qBAAA,CAAsB,oBAAA,EAAsB,GAAA,EAAK,CAAC,CAAA;AAAA,EAC9D;AAEA,EAAA,IAAI,CAAA,KAAM,GAAG,OAAO,CAAA;AACpB,EAAA,IAAI,CAAA,KAAM,GAAG,OAAO,CAAA;AAGpB,EAAA,MAAM,IAAA,GAAO,SAAS,CAAA,GAAI,CAAC,IAAI,QAAA,CAAS,CAAC,IAAI,QAAA,CAAS,CAAC,IAAI,CAAA,GAAI,IAAA,CAAK,IAAI,CAAC,CAAA,GAAI,IAAI,IAAA,CAAK,GAAA,CAAI,IAAI,CAAC,CAAA;AAC/F,EAAA,MAAM,EAAA,GAAK,IAAA,CAAK,GAAA,CAAI,IAAI,CAAA;AAGxB,EAAA,IAAI,CAAA,GAAA,CAAK,CAAA,GAAI,CAAA,KAAM,CAAA,GAAI,IAAI,CAAA,CAAA,EAAI;AAE7B,IAAA,OAAQ,EAAA,GAAK,MAAA,CAAO,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA,GAAK,CAAA;AAAA,EAClC;AAEA,EAAA,OAAO,IAAK,EAAA,GAAK,MAAA,CAAO,GAAG,CAAA,EAAG,CAAA,GAAI,CAAC,CAAA,GAAK,CAAA;AAC1C;AAaA,SAAS,+BAAA,CAAgC,GAAW,CAAA,EAAmB;AAErE,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,KAAK,CAAA,EAAG;AACjC,IAAA,MAAM,IAAI,qBAAA,CAAsB,eAAA,EAAiB,GAAA,EAAK,CAAC,CAAA;AAAA,EACzD;AACA,EAAA,IAAI,IAAI,CAAA,EAAG,MAAM,IAAI,qBAAA,CAAsB,gBAAA,EAAkB,KAAK,CAAC,CAAA;AACnE,EAAA,IAAI,CAAA,KAAM,GAAG,OAAO,CAAA;AAEpB,EAAA,MAAM,KAAA,GAAQ,GAAA;AACd,EAAA,MAAM,GAAA,GAAM,KAAA;AACZ,EAAA,MAAM,KAAA,GAAQ,MAAA;AAEd,EAAA,IAAI,CAAA,GAAI,IAAI,CAAA,EAAG;AAGb,IAAA,IAAI,MAAM,CAAA,GAAI,CAAA;AACd,IAAA,IAAI,GAAA,GAAM,GAAA;AACV,IAAA,IAAI,EAAA,GAAK,CAAA;AAET,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,IAAK,KAAA,EAAO,CAAA,EAAA,EAAK;AAC/B,MAAA,EAAA,IAAM,CAAA;AACN,MAAA,GAAA,IAAO,CAAA,GAAI,EAAA;AACX,MAAA,GAAA,IAAO,GAAA;AACP,MAAA,IAAI,IAAA,CAAK,IAAI,GAAG,CAAA,GAAI,KAAK,GAAA,CAAI,GAAG,IAAI,GAAA,EAAK;AAAA,IAC3C;AAGA,IAAA,OAAO,GAAA,GAAM,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,QAAA,CAAS,CAAC,CAAC,CAAA;AAAA,EAC1D;AAIA,EAAA,IAAI,CAAA,GAAI,IAAI,CAAA,GAAI,CAAA;AAChB,EAAA,IAAI,IAAI,CAAA,GAAI,KAAA;AACZ,EAAA,IAAI,IAAI,CAAA,GAAI,CAAA;AACZ,EAAA,IAAI,CAAA,GAAI,CAAA;AAGR,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,IAAK,KAAA,EAAO,CAAA,EAAA,EAAK;AAC/B,IAAA,MAAM,EAAA,GAAK,CAAC,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA;AACrB,IAAA,CAAA,IAAK,CAAA;AACL,IAAA,CAAA,GAAI,KAAK,CAAA,GAAI,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,IAAI,EAAA,GAAK,CAAA;AACb,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,OAAO,CAAA,GAAI,KAAA;AAC7B,IAAA,CAAA,GAAI,CAAA,GAAI,CAAA;AACR,IAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,IAAA,CAAA,IAAK,GAAA;AACL,IAAA,IAAI,IAAA,CAAK,GAAA,CAAI,GAAA,GAAM,CAAC,IAAI,GAAA,EAAK;AAAA,EAC/B;AAGA,EAAA,MAAM,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAC,CAAA,GAAI,QAAA,CAAS,CAAC,CAAC,CAAA,GAAI,CAAA;AACzD,EAAA,OAAO,CAAA,GAAI,CAAA;AACb;AAkBO,SAAS,UAAU,CAAA,EAAmB;AAG3C,EAAA,MAAM,CAAA,GAAI,KAAA;AACV,EAAA,MAAM,IAAA,GAAO,CAAA,GAAI,CAAA,GAAI,EAAA,GAAK,CAAA;AAC1B,EAAA,MAAM,EAAA,GAAK,IAAA,CAAK,GAAA,CAAI,CAAC,IAAI,IAAA,CAAK,KAAA;AAC9B,EAAA,MAAM,CAAA,GAAI,CAAA,GAAI,CAAA,GAAI,EAAA,GAAK,EAAA;AACvB,EAAA,MAAM,MAAM,IAAA,GAAO,IAAA,CAAK,IAAA,CAAK,CAAA,GAAI,KAAK,GAAA,CAAI,CAAC,EAAA,GAAK,EAAA,IAAA,CAAO,IAAI,IAAA,CAAK,EAAA,GAAK,IAAI,EAAA,GAAK,EAAA,IAAM,EAAE,CAAC,CAAA;AAEvF,EAAA,OAAO,OAAO,CAAA,GAAI,GAAA,CAAA;AACpB;AAmBO,SAAS,WAAA,CAAY,GAAW,EAAA,EAAoB;AAEzD,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,EAAE,CAAA,IAAK,MAAM,CAAA,EAAG;AACnC,IAAA,MAAM,IAAI,qBAAA,CAAsB,gBAAA,EAAkB,IAAA,EAAM,EAAE,CAAA;AAAA,EAC5D;AAEA,EAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,OAAO,GAAA;AAC5B,EAAA,IAAI,CAAC,OAAO,QAAA,CAAS,CAAC,GAAG,OAAO,CAAA,GAAI,IAAI,CAAA,GAAI,CAAA;AAG5C,EAAA,MAAM,CAAA,GAAI,EAAA,IAAM,EAAA,GAAK,CAAA,GAAI,CAAA,CAAA;AACzB,EAAA,MAAM,IAAI,EAAA,GAAK,CAAA;AACf,EAAA,MAAM,CAAA,GAAI,GAAA;AACV,EAAA,MAAM,EAAA,GAAK,yBAAA,CAA0B,CAAA,EAAG,CAAA,EAAG,CAAC,CAAA;AAC5C,EAAA,MAAM,IAAI,GAAA,GAAM,EAAA;AAEhB,EAAA,OAAO,CAAA,IAAK,CAAA,GAAI,CAAA,GAAI,CAAA,GAAI,CAAA;AAC1B;AAmBO,SAAS,YAAA,CAAa,GAAW,CAAA,EAAmB;AAEzD,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,KAAK,CAAA,EAAG;AACjC,IAAA,MAAM,IAAI,qBAAA,CAAsB,gCAAA,EAAkC,GAAA,EAAK,CAAC,CAAA;AAAA,EAC1E;AACA,EAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,OAAO,GAAA;AAC5B,EAAA,IAAI,CAAA,KAAM,UAAU,OAAO,CAAA;AAE3B,EAAA,IAAI,CAAA,IAAK,GAAG,OAAO,CAAA;AAEnB,EAAA,OAAO,+BAAA,CAAgC,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAC,CAAA;AACrD;AAoBO,SAAS,IAAA,CAAK,CAAA,EAAW,GAAA,EAAa,GAAA,EAAqB;AAEhE,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,GAAG,CAAA,IAAK,OAAO,CAAA,EAAG;AACrC,IAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,KAAA,EAAO,GAAG,CAAA;AAAA,EACpF;AACA,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,GAAG,CAAA,IAAK,OAAO,CAAA,EAAG;AACrC,IAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,KAAA,EAAO,GAAG,CAAA;AAAA,EACpF;AACA,EAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,OAAO,GAAA;AAC5B,EAAA,IAAI,CAAA,KAAM,UAAU,OAAO,CAAA;AAE3B,EAAA,IAAI,CAAA,IAAK,GAAG,OAAO,CAAA;AAGnB,EAAA,MAAM,EAAA,GAAM,GAAA,GAAM,CAAA,IAAM,GAAA,GAAM,CAAA,GAAI,GAAA,CAAA;AAClC,EAAA,OAAO,yBAAA,CAA0B,GAAA,GAAM,CAAA,EAAG,GAAA,GAAM,GAAG,EAAE,CAAA;AACvD;;;ACh2BA,SAAS,iBAAiB,CAAA,EAAyB;AACjD,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,CAAA,CAAE,IAAI,CAAA;AACnC,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC3B,IAAA,CAAA,EAAA;AAAA,EACF,CAAC,CAAA;AACD,EAAA,OAAO,GAAA;AACT;AAaA,SAAS,gBAAA,CAAiB,GAAiB,CAAA,EAAyB;AAClE,EAAA,MAAM,IAAI,CAAA,CAAE,MAAA;AAEZ,EAAA,IAAI,IAAA,GAAO,CAAA;AACX,EAAA,IAAI,IAAA,GAAO,CAAA;AACX,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,IAAA,IAAQ,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA;AAChB,IAAA,IAAA,IAAQ,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA;AAAA,EAClB;AACA,EAAA,MAAM,QAAQ,IAAA,GAAO,CAAA;AACrB,EAAA,MAAM,QAAQ,IAAA,GAAO,CAAA;AAGrB,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,IAAA,GAAO,CAAA;AACX,EAAA,IAAI,IAAA,GAAO,CAAA;AACX,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,EAAA,GAAA,CAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACzB,IAAA,MAAM,EAAA,GAAA,CAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACzB,IAAA,GAAA,IAAO,EAAA,GAAK,EAAA;AACZ,IAAA,IAAA,IAAQ,EAAA,GAAK,EAAA;AACb,IAAA,IAAA,IAAQ,EAAA,GAAK,EAAA;AAAA,EACf;AAGA,EAAA,MAAM,GAAA,GAAM,IAAA,CAAK,IAAA,CAAK,IAAA,GAAO,IAAI,CAAA;AACjC,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,4CAAA;AAAA,MACA,OAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,OAAO,GAAA,GAAM,GAAA;AACf;AA0BO,SAAS,QAAA,CAAS,GAAW,CAAA,EAA6B;AAC/D,EAAA,cAAA,CAAe,CAAA,EAAG,GAAG,UAAU,CAAA;AAC/B,EAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,GAAA,EAAK,CAAC,CAAA;AAAA,EACzF;AAEA,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,EAAA,MAAM,CAAA,GAAI,gBAAA,CAAiB,EAAA,EAAI,EAAE,CAAA;AAIjC,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,IAAI,MAAM,CAAA,EAAG;AACX,IAAA,OAAO,CAAC,GAAG,GAAG,CAAA;AAAA,EAChB;AACA,EAAA,MAAM,QAAQ,CAAA,GAAI,IAAA,CAAK,KAAK,EAAA,IAAM,CAAA,GAAI,IAAI,CAAA,CAAE,CAAA;AAC5C,EAAA,MAAM,MAAA,GAAS,KAAK,CAAA,GAAI,WAAA,CAAY,KAAK,GAAA,CAAI,KAAK,GAAG,EAAE,CAAA,CAAA;AACvD,EAAA,OAAO,CAAC,GAAG,MAAM,CAAA;AACnB;AA8BO,SAAS,SAAA,CAAU,GAAW,CAAA,EAA6B;AAChE,EAAA,cAAA,CAAe,CAAA,EAAG,GAAG,WAAW,CAAA;AAChC,EAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,gDAAA,EAAkD,GAAA,EAAK,CAAC,CAAA;AAAA,EAC1F;AAEA,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAG7B,EAAA,MAAM,KAAA,GAAQ,QAAA,CAAS,EAAE,CAAA,CAAE,KAAA;AAC3B,EAAA,MAAM,KAAA,GAAQ,QAAA,CAAS,EAAE,CAAA,CAAE,KAAA;AAG3B,EAAA,MAAM,GAAA,GAAM,gBAAA,CAAiB,KAAA,EAAO,KAAK,CAAA;AAEzC,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,QAAQ,GAAA,GAAM,IAAA,CAAK,KAAK,EAAA,IAAM,CAAA,GAAI,MAAM,GAAA,CAAI,CAAA;AAClD,EAAA,MAAM,MAAA,GAAS,EAAA,GAAK,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,WAAA,CAAY,IAAA,CAAK,GAAA,CAAI,KAAK,CAAA,EAAG,EAAE,CAAA,CAAA,GAAK,GAAA;AACrE,EAAA,OAAO,CAAC,KAAK,MAAM,CAAA;AACrB;AAgCO,SAAS,UAAA,CAAW,GAAW,CAAA,EAA6B;AACjE,EAAA,cAAA,CAAe,CAAA,EAAG,GAAG,YAAY,CAAA;AACjC,EAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,iDAAA,EAAmD,GAAA,EAAK,CAAC,CAAA;AAAA,EAC3F;AAEA,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,EAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAG7B,EAAA,IAAI,UAAA,GAAa,CAAA;AACjB,EAAA,IAAI,UAAA,GAAa,CAAA;AACjB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,GAAI,GAAG,CAAA,EAAA,EAAK;AAC9B,IAAA,MAAM,EAAA,GAAK,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AACpB,IAAA,MAAM,EAAA,GAAK,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AACpB,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,GAAG,CAAA,EAAA,EAAK;AAC9B,MAAA,MAAM,QAAQ,IAAA,CAAK,IAAA,CAAA,CAAM,GAAG,CAAC,CAAA,IAAK,KAAK,EAAE,CAAA;AACzC,MAAA,MAAM,QAAQ,IAAA,CAAK,IAAA,CAAA,CAAM,GAAG,CAAC,CAAA,IAAK,KAAK,EAAE,CAAA;AACzC,MAAA,IAAI,KAAA,KAAU,CAAA,IAAK,KAAA,KAAU,CAAA,EAAG;AAChC,MAAA,IAAI,UAAU,KAAA,EAAO,UAAA,EAAA;AAAA,WAChB,UAAA,EAAA;AAAA,IACP;AAAA,EACF;AAEA,EAAA,MAAM,EAAA,GAAM,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,GAAM,CAAA;AAE3B,EAAA,MAAM,OAAA,GAAU,CACd,IAAA,KAMG;AACH,IAAA,MAAM,MAAA,GAAS,KAAA,CAAM,IAAA,CAAK,IAAI,CAAA,CAAE,KAAK,CAAC,CAAA,EAAG,CAAA,KAAM,CAAA,GAAI,CAAC,CAAA;AACpD,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,MAAA,IAAU;AACnC,MAAA,IAAI,IAAI,CAAA,GAAI,CAAA;AACZ,MAAA,OAAO,CAAA,GAAI,OAAO,MAAA,IAAU,MAAA,CAAO,CAAC,CAAA,KAAM,MAAA,CAAO,CAAC,CAAA,EAAG,CAAA,EAAA;AACrD,MAAA,MAAM,IAAI,CAAA,GAAI,CAAA;AACd,MAAA,IAAI,IAAI,CAAA,EAAG;AACT,QAAA,IAAA,IAAQ,KAAK,CAAA,GAAI,CAAA,CAAA;AACjB,QAAA,KAAA,IAAS,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,CAAA,GAAI,CAAA,CAAA;AAChC,QAAA,KAAA,IAAS,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,CAAA,CAAA;AAAA,MAC9B;AACA,MAAA,CAAA,GAAI,CAAA;AAAA,IACN;AACA,IAAA,OAAO,EAAE,KAAA,EAAO,IAAA,GAAO,CAAA,EAAG,IAAA,EAAM,OAAO,KAAA,EAAM;AAAA,EAC/C,CAAA;AAEA,EAAA,MAAM,IAAA,GAAO,QAAQ,EAAE,CAAA;AACvB,EAAA,MAAM,IAAA,GAAO,QAAQ,EAAE,CAAA;AACvB,EAAA,MAAM,KAAA,GAAQ,KAAK,IAAA,CAAA,CAAM,EAAA,GAAK,KAAK,KAAA,KAAU,EAAA,GAAK,KAAK,KAAA,CAAM,CAAA;AAC7D,EAAA,MAAM,IAAI,UAAA,GAAa,UAAA;AACvB,EAAA,MAAM,GAAA,GAAM,KAAA,KAAU,CAAA,GAAI,GAAA,GAAM,CAAA,GAAI,KAAA;AAIpC,EAAA,IAAI,QACD,CAAA,IAAK,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,GAAI,KAAK,IAAA,CAAK,KAAA,GAAQ,IAAA,CAAK,KAAA,IAAS,KACvD,IAAA,CAAK,IAAA,GAAO,KAAK,IAAA,IAAS,CAAA,GAAI,KAAK,CAAA,GAAI,CAAA,CAAA,CAAA;AAC1C,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,IAAA,IAAS,IAAA,CAAK,QAAQ,IAAA,CAAK,KAAA,IAAU,IAAI,CAAA,IAAK,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,CAAA,CAAA;AAAA,EAC9D;AAEA,EAAA,MAAM,MAAA,GAAS,IAAA,IAAQ,CAAA,GAAI,GAAA,GAAM,KAAK,CAAA,GAAI,SAAA,CAAU,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,IAAA,CAAK,IAAA,CAAK,IAAI,CAAC,CAAC,CAAA,CAAA;AACjF,EAAA,OAAO,CAAC,KAAK,MAAM,CAAA;AACrB;AA0BO,SAAS,QAAA,CAAS,GAAW,CAAA,EAAoB;AACtD,EAAA,IAAI,CAAA,EAAG;AACL,IAAA,MAAM,CAAC,CAAC,CAAA,GAAI,QAAA,CAAS,GAAG,CAAC,CAAA;AACzB,IAAA,OAAO,MAAA,CAAO;AAAA,MACZ,CAAC,GAAK,CAAC,CAAA;AAAA,MACP,CAAC,GAAG,CAAG;AAAA,KACR,CAAA;AAAA,EACH;AAEA,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,IAAI,CAAA,CAAE,OAAO,CAAA,EAAG;AACd,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,6CAAA;AAAA,QACA,MAAA;AAAA,QACA,CAAA,CAAE;AAAA,OACJ;AAAA,IACF;AACA,IAAA,OAAO,MAAA,CAAO,CAAC,CAAC,CAAG,CAAC,CAAC,CAAA;AAAA,EACvB;AAEA,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,WAAW,sCAAsC,CAAA;AAAA,EAC7D;AAGA,EAAA,MAAM,IAAA,GAAO,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,MAAM,IAAA,GAAO,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,IAAI,OAAO,CAAA,EAAG;AACZ,IAAA,MAAM,IAAI,qBAAA,CAAsB,6CAAA,EAA+C,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC7F;AAEA,EAAA,MAAM,EAAA,GAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,MAAM,EAAA,GAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,MAAM,OAAO,CAAA,CAAE,MAAA;AACf,EAAA,MAAM,QAAQ,CAAA,CAAE,IAAA;AAGhB,EAAA,MAAM,eAAe,CAAC,KAAA,CAAM,QAAQ,KAAK,CAAA,IAAK,EAAE,KAAA,YAAiB,aAAA,CAAA;AAEjE,EAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,IAAI,CAAA;AACnC,EAAA,IAAI,YAAA,EAAc;AAChB,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,IAAI,CAAA,GAAI,CAAA;AACR,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,CAAA,IAAK,KAAA,CAAM,IAAA,GAAO,CAAA,GAAI,EAAA,GAAK,IAAI,EAAE,CAAA;AAAA,MACnC;AACA,MAAA,KAAA,CAAM,CAAC,IAAI,CAAA,GAAI,IAAA;AAAA,IACjB;AAAA,EACF,CAAA,MAAO;AACL,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,IAAI,CAAA,GAAI,CAAA;AACR,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,CAAA,IAAK,YAAY,CAAA,EAAG,IAAA,GAAO,CAAA,GAAI,EAAA,GAAK,IAAI,EAAE,CAAA;AAAA,MAC5C;AACA,MAAA,KAAA,CAAM,CAAC,IAAI,CAAA,GAAI,IAAA;AAAA,IACjB;AAAA,EACF;AAEA,EAAA,MAAME,IAAAA,GAAM,IAAI,YAAA,CAAa,IAAA,GAAO,IAAI,CAAA;AACxC,EAAA,MAAM,IAAA,GAAO,CAAA;AACb,EAAA,IAAI,YAAA,EAAc;AAChB,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,EAAA,GAAK,MAAM,CAAC,CAAA;AAClB,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,MAAM,EAAA,GAAK,MAAM,CAAC,CAAA;AAClB,QAAA,IAAI,CAAA,GAAI,CAAA;AACR,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,UAAA,MAAM,IAAA,GAAO,OAAO,CAAA,GAAI,EAAA;AACxB,UAAA,CAAA,IAAA,CAAO,KAAA,CAAM,IAAA,GAAO,CAAA,GAAI,EAAE,CAAA,GAAe,OAAQ,KAAA,CAAM,IAAA,GAAO,CAAA,GAAI,EAAE,CAAA,GAAe,EAAA,CAAA;AAAA,QACrF;AACA,QAAA,MAAM,CAAA,GAAI,KAAK,IAAA,GAAO,IAAA,CAAA;AACtB,QAAAA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AACpB,QAAAA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AAAA,MACtB;AAAA,IACF;AAAA,EACF,CAAA,MAAO;AACL,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,IAAI,CAAA,GAAI,CAAA;AACR,QAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,UAAA,MAAM,IAAA,GAAO,IAAA,GAAO,CAAA,GAAI,EAAA,GAAK,CAAA,GAAI,EAAA;AACjC,UAAA,MAAM,IAAA,GAAO,IAAA,GAAO,CAAA,GAAI,EAAA,GAAK,CAAA,GAAI,EAAA;AACjC,UAAA,CAAA,IAAA,CAAM,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA,IAAK,MAAM,CAAC,CAAA,IAAK,CAAA,CAAA,KAAO,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA,IAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,CAAA,CAAA;AAAA,QACvF;AACA,QAAA,MAAM,CAAA,GAAI,KAAK,IAAA,GAAO,IAAA,CAAA;AACtB,QAAAA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AACpB,QAAAA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AAAA,MACtB;AAAA,IACF;AAAA,EACF;AAEA,EAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,IAAA,GAAO,IAAI,CAAA;AACzC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,CAAA,GAAIA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,IAAK,CAAA;AAC/B,MAAA,MAAM,EAAA,GAAKA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,IAAK,CAAA;AAChC,MAAA,MAAM,EAAA,GAAKA,IAAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,IAAK,CAAA;AAChC,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,IAAA,CAAK,EAAA,GAAK,EAAE,CAAA;AAC7B,MAAA,IAAA,CAAK,IAAI,IAAA,GAAO,CAAC,IAAI,GAAA,KAAQ,CAAA,GAAI,MAAM,CAAA,GAAI,GAAA;AAAA,IAC7C;AAAA,EACF;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,IAAA;AAAA,IACN,KAAA,EAAO,CAAC,IAAA,EAAM,IAAI,CAAA;AAAA,IAClB,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA4BO,SAAS,GAAA,CAAI,CAAA,EAAW,CAAA,EAAY,IAAA,GAAO,CAAA,EAAW;AAC3D,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,IAAI,CAAA,IAAK,OAAO,CAAA,EAAG;AACtC,IAAA,MAAM,IAAI,qBAAA,CAAsB,2CAAA,EAA6C,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC3F;AAEA,EAAA,IAAI,CAAA,EAAG;AACL,IAAA,cAAA,CAAe,CAAA,EAAG,GAAG,KAAK,CAAA;AAC1B,IAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,IAAA,IAAI,MAAM,CAAA,EAAG,MAAM,IAAI,qBAAA,CAAsB,qCAAA,EAAuC,KAAK,CAAC,CAAA;AAC1F,IAAA,IAAI,CAAA,IAAK,IAAA;AACP,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,CAAA,KAAA,EAAQ,IAAI,CAAA,SAAA,EAAY,CAAC,CAAA,sBAAA,CAAA;AAAA,QACzB,MAAA;AAAA,QACA;AAAA,OACF;AAEF,IAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,IAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAE7B,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,KAAA,IAAS,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AAClB,MAAA,KAAA,IAAS,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AAAA,IACpB;AACA,IAAA,KAAA,IAAS,CAAA;AACT,IAAA,KAAA,IAAS,CAAA;AAET,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,EAAA,GAAA,CAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AAC1B,MAAA,MAAM,EAAA,GAAA,CAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AAC1B,MAAA,IAAA,IAAQ,EAAA,GAAK,EAAA;AACb,MAAA,IAAA,IAAQ,EAAA,GAAK,EAAA;AACb,MAAA,KAAA,IAAS,EAAA,GAAK,EAAA;AAAA,IAChB;AACA,IAAA,IAAA,IAAQ,CAAA,GAAI,IAAA;AACZ,IAAA,IAAA,IAAQ,CAAA,GAAI,IAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,IAAA;AAEb,IAAA,OAAO,MAAA,CAAO;AAAA,MACZ,CAAC,MAAM,KAAK,CAAA;AAAA,MACZ,CAAC,OAAO,IAAI;AAAA,KACb,CAAA;AAAA,EACH;AAEA,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,IAAA,IAAI,MAAM,CAAA,EAAG,MAAM,IAAI,qBAAA,CAAsB,qCAAA,EAAuC,KAAK,CAAC,CAAA;AAC1F,IAAA,IAAI,OAAO,CAAA,EAAG;AACZ,MAAA,MAAM,IAAI,qBAAA,CAAsB,2BAAA,EAA6B,MAAA,EAAQ,IAAI,CAAA;AAAA,IAC3E;AACA,IAAA,IAAI,CAAA,IAAK,IAAA;AACP,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,CAAA,KAAA,EAAQ,IAAI,CAAA,SAAA,EAAY,CAAC,CAAA,sBAAA,CAAA;AAAA,QACzB,MAAA;AAAA,QACA;AAAA,OACF;AAEF,IAAA,MAAM,EAAA,GAAK,iBAAiB,CAAC,CAAA;AAC7B,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,EAAG,KAAK,KAAA,IAAS,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AAC9C,IAAA,KAAA,IAAS,CAAA;AACT,IAAA,IAAI,IAAA,GAAO,CAAA;AACX,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,EAAA,GAAA,CAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AAC1B,MAAA,IAAA,IAAQ,EAAA,GAAK,EAAA;AAAA,IACf;AACA,IAAA,IAAA,IAAQ,CAAA,GAAI,IAAA;AACZ,IAAA,OAAO,MAAA,CAAO,CAAC,CAAC,IAAI,CAAC,CAAC,CAAA;AAAA,EACxB;AAEA,EAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,IAAA,MAAM,IAAI,WAAW,iCAAiC,CAAA;AAAA,EACxD;AAEA,EAAA,MAAM,IAAA,GAAO,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,MAAM,IAAA,GAAO,CAAA,CAAE,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAC3B,EAAA,IAAI,IAAA,KAAS,CAAA;AACX,IAAA,MAAM,IAAI,qBAAA,CAAsB,yCAAA,EAA2C,MAAA,EAAQ,IAAI,CAAA;AACzF,EAAA,IAAI,OAAO,CAAA,EAAG;AACZ,IAAA,MAAM,IAAI,qBAAA,CAAsB,2BAAA,EAA6B,MAAA,EAAQ,IAAI,CAAA;AAAA,EAC3E;AACA,EAAA,IAAI,IAAA,IAAQ,IAAA;AACV,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,CAAA,KAAA,EAAQ,IAAI,CAAA,SAAA,EAAY,IAAI,CAAA,sBAAA,CAAA;AAAA,MAC5B,MAAA;AAAA,MACA;AAAA,KACF;AAEF,EAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,IAAI,CAAA;AACnC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,IAAA,IAAI,CAAA,GAAI,CAAA;AACR,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,GAAA,GAAM,CAAA,CAAE,MAAA,GAAS,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA,GAAK,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA;AACtE,MAAA,CAAA,IAAK,WAAA,CAAY,GAAG,GAAG,CAAA;AAAA,IACzB;AACA,IAAA,KAAA,CAAM,CAAC,IAAI,CAAA,GAAI,IAAA;AAAA,EACjB;AAEA,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,IAAA,GAAO,IAAI,CAAA;AACxC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,MAAA,IAAI,CAAA,GAAI,CAAA;AACR,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,EAAM,CAAA,EAAA,EAAK;AAC7B,QAAA,MAAM,IAAA,GAAO,CAAA,CAAE,MAAA,GAAS,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA,GAAK,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA;AACvE,QAAA,MAAM,IAAA,GAAO,CAAA,CAAE,MAAA,GAAS,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA,GAAK,CAAA,IAAK,CAAA,CAAE,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA,CAAA;AACvE,QAAA,CAAA,IAAA,CAAM,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA,IAAK,MAAM,CAAC,CAAA,IAAK,CAAA,CAAA,KAAO,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA,IAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,CAAA,CAAA;AAAA,MACvF;AACA,MAAA,MAAM,CAAA,GAAI,KAAK,IAAA,GAAO,IAAA,CAAA;AACtB,MAAA,GAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AACpB,MAAA,GAAA,CAAI,CAAA,GAAI,IAAA,GAAO,CAAC,CAAA,GAAI,CAAA;AAAA,IACtB;AAAA,EACF;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,CAAC,IAAA,EAAM,IAAI,CAAA;AAAA,IAClB,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;;;AClgBO,SAAS,IAAA,CAAK,CAAA,EAAW,IAAA,EAAiB,SAAA,GAAY,KAAA,EAAe;AAC1E,EAAA,OAAO,UAAA,CAAW,CAAA,EAAG,IAAA,EAAM,SAAS,CAAA;AACtC;AA+BO,SAAS,MAAA,CAAO,CAAA,EAAW,IAAA,EAAiB,SAAA,GAAY,KAAA,EAAe;AAC5E,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AAEvC,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,wCAAA,EAA0C,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,IAC1F;AACA,IAAA,MAAM,SAAmB,EAAC;AAC1B,IAAA,IAAI,MAAA,GAAS,KAAA;AACb,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,MAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,MAAA,GAAS,IAAA;AAC9B,MAAA,MAAA,CAAO,KAAK,CAAC,CAAA;AAAA,IACf,CAAC,CAAA;AACD,IAAA,MAAMH,SAAAA,GAAW,SAAA,GAAY,IAAI,KAAA,CAAc,CAAA,CAAE,IAAI,CAAA,CAAE,IAAA,CAAK,CAAC,CAAA,GAAI,EAAC;AAClE,IAAA,IAAI,MAAA,EAAQ;AACV,MAAA,MAAME,IAAAA,GAAM,IAAI,YAAA,CAAa,CAAC,CAAA;AAC9B,MAAAA,IAAAA,CAAI,CAAC,CAAA,GAAI,MAAA,CAAO,GAAA;AAChB,MAAA,OAAO,OAAO,cAAA,CAAe;AAAA,QAC3B,IAAA,EAAMA,IAAAA;AAAA,QACN,KAAA,EAAOF,SAAAA;AAAA,QACP,KAAA,EAAO,SAAA;AAAA,QACP,QAAQ,CAAA,CAAE;AAAA,OACX,CAAA;AAAA,IACH;AACA,IAAA,MAAA,CAAO,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AAC3B,IAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAM,MAAA,CAAO,SAAS,CAAC,CAAA;AACxC,IAAA,MAAM,SACJ,MAAA,CAAO,MAAA,GAAS,CAAA,KAAM,CAAA,GAAA,CAAA,CAChB,OAAO,GAAA,GAAM,CAAC,CAAA,IAAK,CAAA,KAAM,OAAO,GAAG,CAAA,IAAK,MAAM,CAAA,GAC/C,MAAA,CAAO,GAAG,CAAA,IAAK,CAAA;AACtB,IAAA,MAAME,IAAAA,GAAM,IAAI,YAAA,CAAa,CAAC,CAAA;AAC9B,IAAAA,IAAAA,CAAI,CAAC,CAAA,GAAI,MAAA;AACT,IAAA,OAAO,OAAO,cAAA,CAAe;AAAA,MAC3B,IAAA,EAAMA,IAAAA;AAAA,MACN,KAAA,EAAOF,SAAAA;AAAA,MACP,KAAA,EAAO,SAAA;AAAA,MACP,QAAQ,CAAA,CAAE;AAAA,KACX,CAAA;AAAA,EACH;AAEA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,SAAS,CAAA;AACtD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,OAAA,GAAoC,IAAI,KAAA,CAAM,OAAO,CAAA;AAC3D,EAAA,MAAM,WAAW,IAAI,KAAA,CAAe,OAAO,CAAA,CAAE,KAAK,KAAK,CAAA;AAEvD,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAE3B,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,SAAA,EAAW;AACb,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,MAAM,CAAA,GAAI,UAAA,CAAW,CAAC,CAAA,IAAK,CAAA;AAC3B,QAAA,MAAM,CAAA,GAAI,OAAO,GAAA,CAAI,CAAC,IAAI,CAAA,GAAK,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACzC,QAAA,OAAA,IAAW,CAAA,GAAI,CAAA;AAAA,MACjB;AAAA,IACF,CAAA,MAAO;AACL,MAAA,IAAI,EAAA,GAAK,CAAA;AACT,MAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,QAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,QAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,QAAA,EAAA,EAAA;AAAA,MACF;AAAA,IACF;AAEA,IAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC9B,IAAA,IAAI,MAAA,CAAO,KAAA,CAAM,GAAG,CAAA,EAAG;AACrB,MAAA,QAAA,CAAS,OAAO,CAAA,GAAI,IAAA;AACpB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,OAAO,CAAA,IAAK,EAAC;AACjC,IAAA,GAAA,CAAI,KAAK,GAAG,CAAA;AACZ,IAAA,OAAA,CAAQ,OAAO,CAAA,GAAI,GAAA;AAAA,EACrB,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,IAAI,QAAA,CAAS,CAAC,CAAA,EAAG;AACf,MAAA,GAAA,CAAI,CAAC,IAAI,MAAA,CAAO,GAAA;AAChB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,CAAC,CAAA,IAAK,EAAC;AAC3B,IAAA,IAAI,GAAA,CAAI,WAAW,CAAA,EAAG;AACpB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,iDAAA;AAAA,QACA,MAAA;AAAA,QACA,GAAA,CAAI;AAAA,OACN;AAAA,IACF;AACA,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,IAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAM,GAAA,CAAI,SAAS,CAAC,CAAA;AACrC,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA,CAAI,SAAS,CAAA,KAAM,CAAA,GAAA,CAAA,CAAM,IAAI,GAAA,GAAM,CAAC,CAAA,IAAK,CAAA,KAAM,IAAI,GAAG,CAAA,IAAK,MAAM,CAAA,GAAK,GAAA,CAAI,GAAG,CAAA,IAAK,CAAA;AAAA,EAC7F;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA4BO,SAAS,IAAA,CAAK,GAAW,IAAA,EAAyB;AACvD,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AAEvC,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,IACxF;AACA,IAAA,MAAM,IAAA,uBAAW,GAAA,EAAoB;AACrC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,UAAU,MAAA,CAAO,iBAAA;AACrB,IAAA,IAAI,MAAA,GAAS,KAAA;AACb,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC9B,MAAA,IAAI,MAAA,CAAO,KAAA,CAAM,GAAG,CAAA,EAAG;AACrB,QAAA,MAAA,GAAS,IAAA;AACT,QAAA;AAAA,MACF;AACA,MAAA,MAAM,KAAA,GAAA,CAAS,IAAA,CAAK,GAAA,CAAI,GAAG,KAAK,CAAA,IAAK,CAAA;AACrC,MAAA,IAAA,CAAK,GAAA,CAAI,KAAK,KAAK,CAAA;AACnB,MAAA,IAAI,KAAA,GAAQ,OAAA,IAAY,KAAA,KAAU,OAAA,IAAW,MAAM,OAAA,EAAU;AAC3D,QAAA,OAAA,GAAU,KAAA;AACV,QAAA,OAAA,GAAU,GAAA;AAAA,MACZ;AAAA,IACF,CAAC,CAAA;AACD,IAAA,IAAI,MAAA,EAAQ;AACV,MAAA,OAAO,MAAA,CAAO,CAAC,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,IAC5B;AACA,IAAA,OAAO,MAAA,CAAO,CAAC,OAAO,CAAC,CAAA;AAAA,EACzB;AAGA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,IAAA,GAA+C,IAAI,KAAA,CAAM,OAAO,CAAA;AACtE,EAAA,MAAM,UAAA,GAAa,IAAI,UAAA,CAAW,OAAO,CAAA;AACzC,EAAA,MAAM,UAAA,GAAa,IAAI,YAAA,CAAa,OAAO,CAAA;AAC3C,EAAA,MAAM,WAAW,IAAI,KAAA,CAAe,OAAO,CAAA,CAAE,KAAK,KAAK,CAAA;AACvD,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAE3B,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AAEA,IAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC9B,IAAA,IAAI,MAAA,CAAO,KAAA,CAAM,GAAG,CAAA,EAAG;AACrB,MAAA,QAAA,CAAS,OAAO,CAAA,GAAI,IAAA;AACpB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,CAAA,GAAI,IAAA,CAAK,OAAO,CAAA,wBAAS,GAAA,EAAoB;AACnD,IAAA,MAAM,IAAA,GAAA,CAAQ,CAAA,CAAE,GAAA,CAAI,GAAG,KAAK,CAAA,IAAK,CAAA;AACjC,IAAA,CAAA,CAAE,GAAA,CAAI,KAAK,IAAI,CAAA;AACf,IAAA,IAAA,CAAK,OAAO,CAAA,GAAI,CAAA;AAChB,IAAA,MAAM,gBAAA,GAAmB,UAAA,CAAW,OAAO,CAAA,IAAK,CAAA;AAChD,IAAA,MAAM,gBAAA,GAAmB,UAAA,CAAW,OAAO,CAAA,IAAK,MAAA,CAAO,iBAAA;AACvD,IAAA,IAAI,IAAA,GAAO,gBAAA,IAAqB,IAAA,KAAS,gBAAA,IAAoB,MAAM,gBAAA,EAAmB;AACpF,MAAA,UAAA,CAAW,OAAO,CAAA,GAAI,IAAA;AACtB,MAAA,UAAA,CAAW,OAAO,CAAA,GAAI,GAAA;AAAA,IACxB;AAAA,EACF,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,IAAI,QAAA,CAAS,CAAC,CAAA,EAAG;AACf,MAAA,GAAA,CAAI,CAAC,IAAI,MAAA,CAAO,GAAA;AAChB,MAAA;AAAA,IACF;AACA,IAAA,IAAA,CAAK,UAAA,CAAW,CAAC,CAAA,IAAK,CAAA,MAAO,CAAA,EAAG;AAC9B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,+CAAA;AAAA,QACA,MAAA;AAAA,QACA,UAAA,CAAW,CAAC,CAAA,IAAK;AAAA,OACnB;AAAA,IACF;AACA,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,UAAA,CAAW,CAAC,KAAK,MAAA,CAAO,GAAA;AAAA,EACnC;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA+BO,SAAS,IAAI,CAAA,EAAW,IAAA,EAAiB,SAAA,GAAY,KAAA,EAAO,OAAO,CAAA,EAAW;AACnF,EAAA,MAAM,CAAA,GAAI,cAAA,CAAe,CAAA,EAAG,IAAA,EAAM,WAAW,IAAI,CAAA;AACjD,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,CAAA,CAAE,IAAI,CAAA;AACnC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,IAAA,GAAA,CAAI,CAAC,IAAI,IAAA,CAAK,IAAA,CAAK,YAAY,CAAA,EAAG,CAAA,CAAE,MAAA,GAAS,CAAC,CAAC,CAAA;AAAA,EACjD;AACA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,OAAO,CAAA,CAAE,KAAA;AAAA,IACT,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA+BO,SAAS,SAAS,CAAA,EAAW,IAAA,EAAiB,SAAA,GAAY,KAAA,EAAO,OAAO,CAAA,EAAW;AACxF,EAAA,OAAO,cAAA,CAAe,CAAA,EAAG,IAAA,EAAM,SAAA,EAAW,IAAI,CAAA;AAChD;AAkCO,SAAS,QAAA,CAAS,CAAA,EAAW,IAAA,EAAiB,IAAA,GAAO,IAAA,EAAc;AACxE,EAAA,MAAM,EAAA,GAAK,UAAA,CAAW,CAAA,EAAG,IAAA,EAAM,KAAK,CAAA;AACpC,EAAA,MAAM,MAAA,GAAS,cAAA,CAAe,CAAA,EAAG,IAAA,EAAM,OAAO,CAAC,CAAA;AAE/C,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,OAAA,GAAU,IAAI,YAAA,CAAa,OAAO,CAAA;AACxC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AAEA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,EAAA,EAAI,EAAA,CAAG,SAAS,OAAO,CAAA;AAC7C,IAAA,MAAM,CAAA,GAAI,KAAK,IAAA,CAAK,WAAA,CAAY,QAAQ,MAAA,CAAO,MAAA,GAAS,OAAO,CAAC,CAAA;AAChE,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,MAAM,CAAA,EAAG;AAClC,MAAA,OAAA,CAAQ,OAAO,CAAA,GAAI,GAAA;AAAA,IACrB,CAAA,MAAO;AACL,MAAA,MAAM,CAAA,GAAA,CAAK,IAAI,CAAA,IAAK,CAAA;AACpB,MAAA,OAAA,CAAQ,OAAO,CAAA,GAAA,CAAK,OAAA,CAAQ,OAAO,CAAA,IAAK,CAAA,IAAK,IAAI,CAAA,GAAI,CAAA;AAAA,IACvD;AACA,IAAA,MAAA,CAAO,OAAO,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,KAAK,CAAA,IAAK,CAAA;AAAA,EAC7C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA;AACT,MAAA;AAAA,IACF;AACA,IAAA,IAAI,EAAA,GAAA,CAAM,OAAA,CAAQ,CAAC,CAAA,IAAK,GAAA,IAAO,CAAA;AAC/B,IAAA,IAAI,CAAC,IAAA,EAAM;AAET,MAAA,IAAI,IAAI,CAAA,EAAG;AACT,QAAA,EAAA,GAAK,GAAA;AAAA,MACP,CAAA,MAAO;AACL,QAAA,EAAA,IAAM,KAAK,IAAA,CAAK,CAAA,IAAK,CAAA,GAAI,CAAA,CAAE,KAAK,CAAA,GAAI,CAAA,CAAA;AAAA,MACtC;AAAA,IACF;AACA,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,EAAA;AAAA,EACX;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAiCO,SAAS,SAAS,CAAA,EAAW,IAAA,EAAiB,MAAA,GAAS,IAAA,EAAM,OAAO,IAAA,EAAc;AACvF,EAAA,MAAM,EAAA,GAAK,UAAA,CAAW,CAAA,EAAG,IAAA,EAAM,KAAK,CAAA;AACpC,EAAA,MAAM,MAAA,GAAS,cAAA,CAAe,CAAA,EAAG,IAAA,EAAM,OAAO,CAAC,CAAA;AAE/C,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,OAAA,GAAU,IAAI,YAAA,CAAa,OAAO,CAAA;AACxC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AAEA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,EAAA,EAAI,EAAA,CAAG,SAAS,OAAO,CAAA;AAC7C,IAAA,MAAM,CAAA,GAAI,KAAK,IAAA,CAAK,WAAA,CAAY,QAAQ,MAAA,CAAO,MAAA,GAAS,OAAO,CAAC,CAAA;AAChE,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,MAAM,CAAA,EAAG;AAClC,MAAA,OAAA,CAAQ,OAAO,CAAA,GAAI,GAAA;AAAA,IACrB,CAAA,MAAO;AACL,MAAA,MAAM,CAAA,GAAA,CAAK,IAAI,CAAA,IAAK,CAAA;AACpB,MAAA,OAAA,CAAQ,OAAO,CAAA,GAAA,CAAK,OAAA,CAAQ,OAAO,CAAA,IAAK,KAAK,CAAA,IAAK,CAAA;AAAA,IACpD;AACA,IAAA,MAAA,CAAO,OAAO,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,KAAK,CAAA,IAAK,CAAA;AAAA,EAC7C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA;AACT,MAAA;AAAA,IACF;AACA,IAAA,IAAI,EAAA,GAAA,CAAM,OAAA,CAAQ,CAAC,CAAA,IAAK,GAAA,IAAO,CAAA;AAC/B,IAAA,IAAI,CAAC,IAAA,EAAM;AAET,MAAA,IAAI,IAAI,CAAA,EAAG;AACT,QAAA,EAAA,GAAK,GAAA;AAAA,MACP,CAAA,MAAO;AACL,QAAA,MAAM,SAAS,EAAA,GAAK,CAAA;AACpB,QAAA,MAAM,GAAA,GAAA,CAAA,CAAQ,IAAI,CAAA,IAAK,MAAA,GAAS,OAAO,CAAA,GAAI,CAAA,KAAA,CAAO,CAAA,GAAI,CAAA,KAAM,CAAA,GAAI,CAAA,CAAA,CAAA,CAAA;AAChE,QAAA,EAAA,GAAK,GAAA,GAAM,CAAA;AAAA,MACb;AAAA,IACF;AACA,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,MAAA,GAAS,EAAA,GAAK,CAAA,GAAI,EAAA;AAAA,EAC7B;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAyBO,SAAS,QAAA,CAAS,CAAA,EAAW,CAAA,EAAsB,IAAA,EAAyB;AACjF,EAAA,MAAM,QAAQ,KAAA,CAAM,OAAA,CAAQ,CAAC,CAAA,GAAI,CAAA,GAAI,CAAC,CAAC,CAAA;AACvC,EAAA,KAAA,MAAW,KAAK,KAAA,EAAO;AACrB,IAAA,IAAI,CAAC,OAAO,QAAA,CAAS,CAAC,KAAK,CAAA,GAAI,CAAA,IAAK,IAAI,CAAA,EAAG;AACzC,MAAA,MAAM,IAAI,qBAAA,CAAsB,qBAAA,EAAuB,GAAA,EAAK,CAAC,CAAA;AAAA,IAC/D;AAAA,EACF;AAEA,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAE3B,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,IAAI,CAAA,CAAE,SAAS,CAAA,EAAG;AAChB,MAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,IAC5F;AACA,IAAA,MAAM,MAAgB,EAAC;AACvB,IAAA,IAAI,MAAA,GAAS,KAAA;AACb,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,MAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,MAAA,GAAS,IAAA;AAC9B,MAAA,GAAA,CAAI,KAAK,CAAC,CAAA;AAAA,IACZ,CAAC,CAAA;AACD,IAAA,IAAI,MAAA,EAAQ;AACV,MAAA,OAAO,OAAO,KAAA,CAAM,GAAA,CAAI,MAAM,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,IAC3C;AACA,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,IAAA,MAAM,UAAoB,EAAC;AAC3B,IAAA,KAAA,MAAW,QAAQ,KAAA,EAAO;AACxB,MAAA,MAAM,GAAA,GAAM,IAAA,IAAQ,GAAA,CAAI,MAAA,GAAS,CAAA,CAAA;AACjC,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,GAAG,CAAA;AAC5B,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,IAAA,CAAK,GAAG,CAAA;AAC3B,MAAA,MAAM,SAAS,GAAA,GAAM,KAAA;AACrB,MAAA,OAAA,CAAQ,IAAA,CAAA,CAAM,GAAA,CAAI,KAAK,CAAA,IAAK,CAAA,KAAM,CAAA,GAAI,MAAA,CAAA,GAAA,CAAW,GAAA,CAAI,KAAK,CAAA,IAAK,CAAA,IAAK,MAAM,CAAA;AAAA,IAC5E;AACA,IAAA,OAAO,OAAO,OAAO,CAAA;AAAA,EACvB;AAEA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,OAAA,GAAoC,IAAI,KAAA,CAAM,OAAO,CAAA;AAC3D,EAAA,MAAM,WAAW,IAAI,KAAA,CAAe,OAAO,CAAA,CAAE,KAAK,KAAK,CAAA;AAEvD,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC9B,IAAA,IAAI,MAAA,CAAO,KAAA,CAAM,GAAG,CAAA,EAAG;AACrB,MAAA,QAAA,CAAS,OAAO,CAAA,GAAI,IAAA;AACpB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,OAAO,CAAA,IAAK,EAAC;AACjC,IAAA,GAAA,CAAI,KAAK,GAAG,CAAA;AACZ,IAAA,OAAA,CAAQ,OAAO,CAAA,GAAI,GAAA;AAAA,EACrB,CAAC,CAAA;AAGD,EAAA,MAAM,UAAA,GAAa,CAAC,KAAA,CAAM,MAAA,EAAQ,GAAG,QAAQ,CAAA;AAC7C,EAAA,MAAM,SAAA,GAAY,MAAM,MAAA,GAAS,OAAA;AACjC,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,SAAS,CAAA;AAEtC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,IAAI,QAAA,CAAS,CAAC,CAAA,EAAG;AACf,MAAA,KAAA,IAAS,EAAA,GAAK,CAAA,EAAG,EAAA,GAAK,KAAA,CAAM,QAAQ,EAAA,EAAA,EAAM;AACxC,QAAA,GAAA,CAAI,EAAA,GAAK,OAAA,GAAU,CAAC,CAAA,GAAI,MAAA,CAAO,GAAA;AAAA,MACjC;AACA,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,CAAC,CAAA,IAAK,EAAC;AAC3B,IAAA,IAAI,GAAA,CAAI,WAAW,CAAA,EAAG;AACpB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,mDAAA;AAAA,QACA,MAAA;AAAA,QACA,GAAA,CAAI;AAAA,OACN;AAAA,IACF;AACA,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,IAAA,KAAA,IAAS,EAAA,GAAK,CAAA,EAAG,EAAA,GAAK,KAAA,CAAM,QAAQ,EAAA,EAAA,EAAM;AACxC,MAAA,MAAM,IAAA,GAAO,KAAA,CAAM,EAAE,CAAA,IAAK,CAAA;AAC1B,MAAA,MAAM,GAAA,GAAM,IAAA,IAAQ,GAAA,CAAI,MAAA,GAAS,CAAA,CAAA;AACjC,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,GAAG,CAAA;AAC5B,MAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,IAAA,CAAK,GAAG,CAAA;AAC3B,MAAA,MAAM,SAAS,GAAA,GAAM,KAAA;AACrB,MAAA,GAAA,CAAI,EAAA,GAAK,OAAA,GAAU,CAAC,CAAA,GAAA,CAAK,GAAA,CAAI,KAAK,CAAA,IAAK,CAAA,KAAM,CAAA,GAAI,MAAA,CAAA,GAAA,CAAW,GAAA,CAAI,KAAK,KAAK,CAAA,IAAK,MAAA;AAAA,IACjF;AAAA,EACF;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,UAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAyBO,SAAS,UAAA,CAAW,CAAA,EAAW,CAAA,EAAsB,IAAA,EAAyB;AACnF,EAAA,MAAM,OAAO,KAAA,CAAM,OAAA,CAAQ,CAAC,CAAA,GAAI,CAAA,GAAI,CAAC,CAAC,CAAA;AACtC,EAAA,KAAA,MAAW,KAAK,IAAA,EAAM;AACpB,IAAA,IAAI,CAAC,OAAO,QAAA,CAAS,CAAC,KAAK,CAAA,GAAI,CAAA,IAAK,IAAI,GAAA,EAAK;AAC3C,MAAA,MAAM,IAAI,qBAAA,CAAsB,uBAAA,EAAyB,GAAA,EAAK,CAAC,CAAA;AAAA,IACjE;AAAA,EACF;AAEA,EAAA,MAAM,KAAA,GAAQ,KAAA,CAAM,OAAA,CAAQ,CAAC,CAAA,GAAI,CAAA,CAAE,GAAA,CAAI,CAAC,CAAA,KAAM,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,GAAI,GAAA;AAC7D,EAAA,OAAO,QAAA,CAAS,CAAA,EAAG,KAAA,EAAO,IAAI,CAAA;AAChC;AA4BO,SAAS,MAAA,CAAO,CAAA,EAAW,CAAA,EAAW,IAAA,EAAyB;AACpE,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,CAAC,MAAA,CAAO,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA,GAAI,CAAA,EAAG;AACxD,IAAA,MAAM,IAAI,qBAAA,CAAsB,kCAAA,EAAoC,GAAA,EAAK,CAAC,CAAA;AAAA,EAC5E;AAEA,EAAA,MAAM,EAAA,GAAK,UAAA,CAAW,CAAA,EAAG,IAAA,EAAM,KAAK,CAAA;AACpC,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,OAAO,CAAA;AACrC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,EAAA,EAAI,EAAA,CAAG,SAAS,OAAO,CAAA;AAC7C,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,IAAA,CAAK,OAAO,CAAA,GAAA,CAAK,IAAA,CAAK,OAAO,CAAA,IAAK,CAAA,IAAA,CAAM,IAAI,CAAA,KAAM,CAAA;AAClD,IAAA,MAAA,CAAO,OAAO,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,KAAK,CAAA,IAAK,CAAA;AAAA,EAC7C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,GAAA,CAAI,CAAC,IAAI,CAAA,KAAM,CAAA,GAAI,OAAO,IAAA,CAAK,CAAC,KAAK,CAAA,IAAK,CAAA;AAAA,EAC5C;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA2BO,SAAS,aAAA,CAAc,GAAW,IAAA,EAAyB;AAChE,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,IAAI,IAAA,CAAK,MAAA,KAAW,CAAA,IAAK,CAAA,CAAE,SAAS,CAAA,EAAG;AACrC,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,+CAAA;AAAA,MACA,MAAA;AAAA,MACA,CAAA,CAAE;AAAA,KACJ;AAAA,EACF;AACA,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,OAAO,CAAA;AACrC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,IAAI,CAAA,IAAK,CAAA;AACP,MAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,OAAA,EAAS,CAAC,CAAA;AAC7F,IAAA,IAAA,CAAK,OAAO,KAAK,IAAA,CAAK,OAAO,KAAK,CAAA,IAAK,IAAA,CAAK,IAAI,CAAC,CAAA;AACjD,IAAA,MAAA,CAAO,OAAO,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,KAAK,CAAA,IAAK,CAAA;AAAA,EAC7C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,wDAAA;AAAA,QACA,MAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,GAAA,CAAI,CAAC,IAAI,IAAA,CAAK,GAAA,CAAA,CAAK,KAAK,CAAC,CAAA,IAAK,KAAK,CAAC,CAAA;AAAA,EACtC;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AA2BO,SAAS,YAAA,CAAa,GAAW,IAAA,EAAyB;AAC/D,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,IAAI,IAAA,CAAK,MAAA,KAAW,CAAA,IAAK,CAAA,CAAE,SAAS,CAAA,EAAG;AACrC,IAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,MAAA,EAAQ,EAAE,IAAI,CAAA;AAAA,EAChG;AACA,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAC3B,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,OAAO,CAAA;AACrC,EAAA,MAAM,MAAA,GAAS,IAAI,UAAA,CAAW,OAAO,CAAA;AAErC,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,IAAA,IAAI,KAAK,CAAA,EAAG;AACV,MAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,OAAA,EAAS,CAAC,CAAA;AAAA,IAC5F;AACA,IAAA,IAAA,CAAK,OAAO,CAAA,GAAA,CAAK,IAAA,CAAK,OAAO,CAAA,IAAK,KAAK,CAAA,GAAI,CAAA;AAC3C,IAAA,MAAA,CAAO,OAAO,CAAA,GAAA,CAAK,MAAA,CAAO,OAAO,KAAK,CAAA,IAAK,CAAA;AAAA,EAC7C,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,MAAM,CAAA,GAAI,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,uDAAA;AAAA,QACA,MAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,CAAA,IAAK,IAAA,CAAK,CAAC,CAAA,IAAK,GAAA,CAAA;AAAA,EAC3B;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;AAyBO,SAAS,QAAA,CAAS,CAAA,EAAW,eAAA,EAAyB,IAAA,EAAyB;AACpF,EAAA,IAAI,CAAC,OAAO,QAAA,CAAS,eAAe,KAAK,eAAA,GAAkB,CAAA,IAAK,mBAAmB,GAAA,EAAK;AACtF,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,2DAAA;AAAA,MACA,iBAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,MAAM,IAAA,GAAO,aAAA,CAAc,IAAA,EAAM,CAAA,CAAE,IAAI,CAAA;AACvC,EAAA,MAAM,MAAA,GAAS,IAAI,GAAA,CAAI,IAAI,CAAA;AAE3B,EAAA,IAAI,IAAA,CAAK,WAAW,CAAA,EAAG;AACrB,IAAA,MAAM,MAAgB,EAAC;AACvB,IAAA,IAAI,MAAA,GAAS,KAAA;AACb,IAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,MAAA,MAAM,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC5B,MAAA,IAAI,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,EAAG,MAAA,GAAS,IAAA;AAC9B,MAAA,GAAA,CAAI,KAAK,CAAC,CAAA;AAAA,IACZ,CAAC,CAAA;AACD,IAAA,IAAI,IAAI,MAAA,KAAW,CAAA;AACjB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,0CAAA;AAAA,QACA,MAAA;AAAA,QACA,GAAA,CAAI;AAAA,OACN;AACF,IAAA,IAAI,MAAA,EAAQ;AACV,MAAA,OAAO,MAAA,CAAO,CAAC,MAAA,CAAO,GAAG,CAAC,CAAA;AAAA,IAC5B;AACA,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,GAAA,CAAI,SAAS,eAAe,CAAA;AACrD,IAAA,MAAM,UAAU,GAAA,CAAI,KAAA,CAAM,KAAA,EAAO,GAAA,CAAI,SAAS,KAAK,CAAA;AACnD,IAAA,MAAM,GAAA,GAAM,QAAQ,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAC7C,IAAA,OAAO,MAAA,CAAO,CAAC,GAAA,GAAM,OAAA,CAAQ,MAAM,CAAC,CAAA;AAAA,EACtC;AAEA,EAAA,MAAM,QAAA,GAAW,YAAA,CAAa,CAAA,CAAE,KAAA,EAAO,MAAM,KAAK,CAAA;AAClD,EAAA,MAAM,UAAA,GAAa,eAAe,QAAQ,CAAA;AAC1C,EAAA,MAAM,OAAA,GAAU,SAAS,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAClD,EAAA,MAAM,OAAA,GAAoC,IAAI,KAAA,CAAM,OAAO,CAAA;AAC3D,EAAA,MAAM,WAAW,IAAI,KAAA,CAAe,OAAO,CAAA,CAAE,KAAK,KAAK,CAAA;AAEvD,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,EAAK,GAAA,KAAQ;AAClC,IAAA,IAAI,OAAA,GAAU,CAAA;AACd,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,MAAM,CAAA,EAAA,EAAK;AAC/B,MAAA,IAAI,MAAA,CAAO,GAAA,CAAI,CAAC,CAAA,EAAG;AACnB,MAAA,OAAA,IAAA,CAAY,IAAI,CAAC,CAAA,IAAK,CAAA,KAAM,UAAA,CAAW,EAAE,CAAA,IAAK,CAAA,CAAA;AAC9C,MAAA,EAAA,EAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC9B,IAAA,IAAI,MAAA,CAAO,KAAA,CAAM,GAAG,CAAA,EAAG;AACrB,MAAA,QAAA,CAAS,OAAO,CAAA,GAAI,IAAA;AACpB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,OAAO,CAAA,IAAK,EAAC;AACjC,IAAA,GAAA,CAAI,KAAK,GAAG,CAAA;AACZ,IAAA,OAAA,CAAQ,OAAO,CAAA,GAAI,GAAA;AAAA,EACrB,CAAC,CAAA;AAED,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,OAAO,CAAA;AACpC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,OAAA,EAAS,CAAA,EAAA,EAAK;AAChC,IAAA,IAAI,QAAA,CAAS,CAAC,CAAA,EAAG;AACf,MAAA,GAAA,CAAI,CAAC,IAAI,MAAA,CAAO,GAAA;AAChB,MAAA;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,OAAA,CAAQ,CAAC,CAAA,IAAK,EAAC;AAC3B,IAAA,IAAI,GAAA,CAAI,WAAW,CAAA,EAAG;AACpB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,mDAAA;AAAA,QACA,MAAA;AAAA,QACA,GAAA,CAAI;AAAA,OACN;AAAA,IACF;AACA,IAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,KAAA,CAAM,GAAA,CAAI,SAAS,eAAe,CAAA;AACrD,IAAA,MAAM,UAAU,GAAA,CAAI,KAAA,CAAM,KAAA,EAAO,GAAA,CAAI,SAAS,KAAK,CAAA;AACnD,IAAA,MAAM,GAAA,GAAM,QAAQ,MAAA,CAAO,CAAC,GAAG,CAAA,KAAM,CAAA,GAAI,GAAG,CAAC,CAAA;AAC7C,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,GAAA,GAAM,OAAA,CAAQ,MAAA;AAAA,EACzB;AAEA,EAAA,OAAO,OAAO,cAAA,CAAe;AAAA,IAC3B,IAAA,EAAM,GAAA;AAAA,IACN,KAAA,EAAO,QAAA;AAAA,IACP,KAAA,EAAO,SAAA;AAAA,IACP,QAAQ,CAAA,CAAE;AAAA,GACX,CAAA;AACH;;;AClgCA,SAAS,qBAAqB,CAAA,EAAyB;AACrD,EAAA,IAAI,CAAA,CAAE,OAAO,CAAA,EAAG;AACd,IAAA,OAAO,IAAI,aAAa,CAAC,CAAA;AAAA,EAC3B;AACA,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,CAAA,CAAE,IAAI,CAAA;AACnC,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC3B,IAAA,CAAA,EAAA;AAAA,EACF,CAAC,CAAA;AACD,EAAA,GAAA,CAAI,IAAA,CAAK,CAAC,CAAA,EAAG,CAAA,KAAM,IAAI,CAAC,CAAA;AACxB,EAAA,OAAO,GAAA;AACT;AAEA,SAAS,eAAe,CAAA,EAAyB;AAC/C,EAAA,MAAM,GAAA,GAAM,IAAI,YAAA,CAAa,CAAA,CAAE,IAAI,CAAA;AACnC,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,GAAA,KAAQ;AAC7B,IAAA,GAAA,CAAI,CAAC,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,GAAG,CAAA;AAC3B,IAAA,CAAA,EAAA;AAAA,EACF,CAAC,CAAA;AACD,EAAA,OAAO,GAAA;AACT;AAEA,SAAS,UAAU,CAAA,EAA+C;AAChE,EAAA,IAAI,CAAA,CAAE,WAAW,CAAA,EAAG;AAClB,IAAA,MAAM,IAAI,qBAAA,CAAsB,+BAAA,EAAiC,QAAA,EAAU,EAAE,MAAM,CAAA;AAAA,EACrF;AACA,EAAA,IAAIC,KAAAA,GAAO,CAAA;AACX,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,QAAQ,CAAA,EAAA,EAAK;AACjC,IAAA,MAAM,CAAA,GAAI,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA;AAClB,IAAA,MAAM,IAAI,CAAA,GAAI,CAAA;AACd,IAAA,MAAM,QAAQ,CAAA,GAAIA,KAAAA;AAClB,IAAAA,SAAQ,KAAA,GAAQ,CAAA;AAChB,IAAA,MAAM,SAAS,CAAA,GAAIA,KAAAA;AACnB,IAAA,EAAA,IAAM,KAAA,GAAQ,MAAA;AAAA,EAChB;AACA,EAAA,OAAO,EAAE,IAAA,EAAAA,KAAAA,EAAM,EAAA,EAAG;AACpB;AAEA,SAAS,YAAY,CAAA,EAA6B;AAGhD,EAAA,MAAM,IAAI,CAAA,CAAE,MAAA;AACZ,EAAA,IAAI,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,GAAA,EAAM;AACrB,IAAA,MAAM,IAAI,qBAAA,CAAsB,kDAAA,EAAoD,GAAA,EAAK,CAAC,CAAA;AAAA,EAC5F;AAEA,EAAA,MAAM,KAAA,GAAA,CAAS,EAAE,CAAA,GAAI,CAAC,KAAK,CAAA,KAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA;AACzC,EAAA,IAAI,UAAU,CAAA,EAAG;AACf,IAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,OAAA,EAAS,KAAK,CAAA;AAAA,EACxF;AAEA,EAAA,MAAM,KAAA,GAAQ,KAAA;AACd,EAAA,IAAI,QAAQ,KAAA,EAAO;AACjB,IAAA,MAAM,IAAI,qBAAA,CAAsB,8BAAA,EAAgC,OAAA,EAAS,KAAK,CAAA;AAAA,EAChF;AAEA,EAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAM,CAAA,GAAI,CAAC,CAAA;AAC5B,EAAA,MAAM,CAAA,GAAI,IAAI,YAAA,CAAa,GAAA,GAAM,CAAC,CAAA;AAElC,EAAA,MAAM,CAAA,GAAuB,CAAC,MAAA,EAAQ,KAAK,CAAA;AAC3C,EAAA,MAAM,KAAwB,CAAC,CAAA,EAAK,UAAU,SAAA,EAAW,QAAA,EAAU,UAAU,SAAS,CAAA;AACtF,EAAA,MAAM,KAAwB,CAAC,CAAA,EAAK,UAAU,SAAA,EAAW,SAAA,EAAW,UAAU,SAAS,CAAA;AACvF,EAAA,MAAM,EAAA,GAAwB,CAAC,KAAA,EAAO,QAAA,EAAU,UAAU,QAAS,CAAA;AACnE,EAAA,MAAM,EAAA,GAAwB,CAAC,MAAA,EAAQ,QAAA,EAAU,UAAU,SAAU,CAAA;AACrE,EAAA,MAAM,EAAA,GAAwB,CAAC,OAAA,EAAS,QAAA,EAAU,WAAW,QAAS,CAAA;AACtE,EAAA,MAAM,EAAA,GAAwB,CAAC,OAAA,EAAS,SAAA,EAAW,QAAS,CAAA;AAE5D,EAAA,MAAM,IAAA,GAAO,CAAC,EAAA,EAAuB,EAAA,KAAuB;AAC1D,IAAA,IAAI,CAAA,GAAI,EAAA,CAAG,EAAA,CAAG,MAAA,GAAS,CAAC,CAAA,IAAK,CAAA;AAC7B,IAAA,KAAA,IAAS,IAAI,EAAA,CAAG,MAAA,GAAS,CAAA,EAAG,CAAA,IAAK,GAAG,CAAA,EAAA,EAAK;AACvC,MAAA,CAAA,GAAI,CAAA,GAAI,EAAA,IAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,CAAA;AAAA,IACzB;AACA,IAAA,OAAO,CAAA;AAAA,EACT,CAAA;AAEA,EAAA,MAAM,IAAA,GAAO,CAAC,CAAA,KAAuB,CAAA,KAAM,IAAI,CAAA,GAAI,CAAA,GAAI,IAAI,CAAA,GAAI,EAAA;AAE/D,EAAA,MAAM,EAAA,GAAK,CAAA;AACX,EAAA,IAAI,MAAM,CAAA,EAAG;AACX,IAAA,CAAA,CAAE,CAAC,IAAI,IAAA,CAAK,OAAA;AAAA,EACd,CAAA,MAAO;AACL,IAAA,MAAM,OAAO,EAAA,GAAK,IAAA;AAClB,IAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,IAAK,GAAA,EAAK,CAAA,EAAA,EAAK;AAC7B,MAAA,MAAM,CAAA,GAAA,CAAK,IAAI,KAAA,IAAS,IAAA;AAGxB,MAAA,IAAI,EAAA,GAAK,GAAA;AACT,MAAA,IAAI,EAAA,GAAK,EAAA;AACT,MAAA,KAAA,IAAS,EAAA,GAAK,CAAA,EAAG,EAAA,GAAK,EAAA,EAAI,EAAA,EAAA,EAAM;AAC9B,QAAA,MAAM,GAAA,GAAA,CAAO,KAAK,EAAA,IAAM,CAAA;AACxB,QAAA,MAAM,GAAA,GAAM,UAAU,GAAG,CAAA;AACzB,QAAA,IAAI,GAAA,GAAM,GAAG,EAAA,GAAK,GAAA;AAAA,aACb,EAAA,GAAK,GAAA;AAAA,MACZ;AACA,MAAA,MAAMG,EAAAA,GAAAA,CAAK,KAAK,EAAA,IAAM,CAAA;AACtB,MAAA,CAAA,CAAE,CAAC,CAAA,GAAIA,EAAAA;AACP,MAAA,KAAA,IAASA,EAAAA,GAAIA,EAAAA;AAAA,IACf;AACA,IAAA,KAAA,IAAS,CAAA;AACT,IAAA,MAAM,MAAA,GAAS,IAAA,CAAK,IAAA,CAAK,KAAK,CAAA;AAC9B,IAAA,MAAM,GAAA,GAAM,CAAA,GAAI,IAAA,CAAK,IAAA,CAAK,EAAE,CAAA;AAE5B,IAAA,MAAM,EAAA,GAAK,KAAK,EAAA,EAAI,GAAG,KAAK,CAAA,CAAE,CAAC,KAAK,CAAA,IAAK,MAAA;AACzC,IAAA,IAAI,EAAA;AACJ,IAAA,IAAI,GAAA;AACJ,IAAA,IAAI,IAAI,CAAA,EAAG;AACT,MAAA,EAAA,GAAK,CAAA;AACL,MAAA,MAAM,EAAA,GAAK,GAAG,CAAA,CAAE,CAAC,KAAK,CAAA,IAAK,MAAA,CAAA,GAAU,IAAA,CAAK,EAAA,EAAI,GAAG,CAAA;AACjD,MAAA,GAAA,GAAM,IAAA,CAAK,IAAA;AAAA,QAAA,CACR,KAAA,GAAQ,CAAA,IAAK,CAAA,CAAE,CAAC,CAAA,IAAK,MAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA,GAAK,CAAA,IAAK,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA,IAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA,KACjE,IAAI,CAAA,GAAI,EAAA,GAAK,EAAA,GAAK,CAAA,GAAI,EAAA,GAAK,EAAA;AAAA,OAChC;AACA,MAAA,CAAA,CAAE,CAAC,CAAA,GAAI,EAAA;AAAA,IACT,CAAA,MAAO;AACL,MAAA,EAAA,GAAK,CAAA;AACL,MAAA,GAAA,GAAM,IAAA,CAAK,IAAA,CAAA,CAAM,KAAA,GAAQ,CAAA,IAAK,EAAE,CAAC,CAAA,IAAK,CAAA,CAAA,IAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA,KAAO,CAAA,GAAI,CAAA,GAAI,KAAK,EAAA,CAAG,CAAA;AAAA,IAC7E;AACA,IAAA,CAAA,CAAE,CAAC,CAAA,GAAI,EAAA;AACP,IAAA,KAAA,IAAS,CAAA,GAAI,EAAA,EAAI,CAAA,IAAK,GAAA,EAAK,CAAA,EAAA,EAAK;AAC9B,MAAA,CAAA,CAAE,CAAC,CAAA,GAAI,EAAA,CAAG,CAAA,CAAE,CAAC,KAAK,CAAA,IAAK,GAAA,CAAA;AAAA,IACzB;AAAA,EACF;AAGA,EAAA,IAAI,EAAA,GAAA,CAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACvB,EAAA,IAAI,EAAA,GAAK,EAAA;AACT,EAAA,IAAI,EAAA,GAAK,EAAE,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,CAAA;AACnB,EAAA,KAAA,IAAS,IAAI,CAAA,EAAG,CAAA,GAAI,IAAI,CAAA,EAAG,CAAA,GAAI,GAAG,CAAA,EAAA,EAAK;AACrC,IAAA,MAAM,EAAA,GAAA,CAAM,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACzB,IAAA,IAAI,EAAA,GAAK,KAAK,KAAA,EAAO;AACnB,MAAA,MAAM,IAAI,qBAAA,CAAsB,8BAAA,EAAgC,MAAA,EAAQ,UAAU,CAAA;AAAA,IACpF;AACA,IAAA,EAAA,IAAM,EAAA;AACN,IAAA,CAAA,EAAA;AACA,IAAA,IAAI,MAAM,CAAA,EAAG;AACX,MAAA,EAAA,IAAM,IAAA,CAAK,CAAA,GAAI,CAAC,CAAA,IAAK,CAAA,CAAE,KAAK,GAAA,CAAI,CAAA,EAAG,CAAC,CAAC,CAAA,IAAK,CAAA,CAAA;AAAA,IAC5C;AACA,IAAA,EAAA,GAAK,EAAA;AAAA,EACP;AAEA,EAAA,EAAA,IAAM,CAAA;AACN,EAAA,EAAA,IAAM,CAAA;AAEN,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,EAAG,KAAK,CAAA,EAAA,EAAK;AAC1C,IAAA,MAAM,MAAM,CAAA,KAAM,CAAA,GAAI,IAAA,CAAK,CAAA,GAAI,CAAC,CAAA,IAAK,CAAA,CAAE,CAAA,GAAI,IAAA,CAAK,IAAI,CAAA,EAAG,CAAC,CAAC,CAAA,IAAK,CAAA,CAAA,GAAK,KAAK,CAAC,EAAA;AACzE,IAAA,MAAM,GAAA,GAAA,CAAO,CAAA,CAAE,CAAC,CAAA,IAAK,KAAK,KAAA,GAAQ,EAAA;AAClC,IAAA,GAAA,IAAO,GAAA,GAAM,GAAA;AACb,IAAA,GAAA,IAAO,GAAA,GAAM,GAAA;AACb,IAAA,GAAA,IAAO,GAAA,GAAM,GAAA;AAAA,EACf;AAEA,EAAA,MAAM,MAAA,GAAS,IAAA,CAAK,IAAA,CAAK,GAAA,GAAM,GAAG,CAAA;AAClC,EAAA,MAAM,EAAA,GAAA,CAAO,MAAA,GAAS,GAAA,KAAQ,MAAA,GAAS,QAAS,GAAA,GAAM,GAAA,CAAA;AACtD,EAAA,MAAM,IAAI,CAAA,GAAI,EAAA;AAEd,EAAA,IAAI,MAAM,CAAA,EAAG;AACX,IAAA,MAAM,GAAA,GAAM,gBAAA;AACZ,IAAA,MAAM,IAAA,GAAO,eAAA;AACb,IAAA,IAAI,EAAA,GAAK,OAAO,IAAA,CAAK,IAAA,CAAK,KAAK,IAAA,CAAK,CAAC,CAAC,CAAA,GAAI,IAAA,CAAA;AAC1C,IAAA,IAAI,EAAA,GAAK,GAAG,EAAA,GAAK,CAAA;AACjB,IAAA,IAAI,EAAA,GAAK,GAAG,EAAA,GAAK,CAAA;AACjB,IAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,EAAA,EAAG;AAAA,EACpC;AAEA,EAAA,MAAM,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,EAAE,CAAA;AACrB,EAAA,MAAM,GAAA,GAAM,IAAA,CAAK,GAAA,CAAI,EAAE,CAAA;AACvB,EAAA,IAAI,CAAA;AACJ,EAAA,IAAI,CAAA;AACJ,EAAA,IAAI,KAAK,EAAA,EAAI;AACX,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,CAAA,EAAG,EAAE,CAAA;AACxB,IAAA,IAAI,KAAK,KAAA,EAAO;AACd,MAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,CAAA,EAAE;AAAA,IACnC;AACA,IAAA,MAAM,EAAA,GAAK,CAAC,IAAA,CAAK,GAAA,CAAI,QAAQ,CAAC,CAAA;AAC9B,IAAA,CAAA,GAAI,IAAA,CAAK,IAAI,EAAE,CAAA;AACf,IAAA,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,EAAA,EAAI,EAAE,CAAC,CAAA;AACzB,IAAA,MAAMA,EAAAA,GAAAA,CAAK,KAAK,CAAA,IAAK,CAAA;AACrB,IAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,SAAA,CAAUA,EAAC,CAAA,EAAE;AAAA,EAC9C;AACA,EAAA,CAAA,GAAI,IAAA,CAAK,IAAI,GAAG,CAAA;AAChB,EAAA,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,EAAA,EAAI,GAAG,CAAC,CAAA;AAC1B,EAAA,MAAM,CAAA,GAAA,CAAK,IAAI,CAAA,IAAK,CAAA;AACpB,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,SAAA,CAAU,CAAC,CAAA,EAAE;AAC9C;AASO,SAAS,WAAA,CAAY,GAAW,OAAA,EAA6B;AAClE,EAAA,MAAM,CAAA,GAAI,qBAAqB,CAAC,CAAA;AAChC,EAAA,MAAM,IAAI,CAAA,CAAE,MAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,2CAAA,EAA6C,GAAA,EAAK,CAAC,CAAA;AAAA,EACrF;AAGA,EAAA,IAAIH,KAAAA,GAAO,CAAA;AACX,EAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,EAAG,KAAKA,KAAAA,IAAQ,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA;AAC5C,EAAAA,KAAAA,IAAQ,CAAA;AAGR,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,CAAA,GAAA,CAAK,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAKA,KAAAA;AACxB,IAAA,EAAA,IAAM,CAAA,GAAI,CAAA;AAAA,EACZ;AACA,EAAA,MAAMI,SAAAA,GAAW,MAAM,CAAA,GAAI,CAAA,CAAA;AAC3B,EAAA,MAAMC,IAAAA,GAAM,IAAA,CAAK,IAAA,CAAKD,SAAQ,CAAA;AAC9B,EAAA,IAAIC,SAAQ,CAAA,EAAG;AACb,IAAA,MAAM,IAAI,qBAAA,CAAsB,+CAAA,EAAiD,KAAA,EAAOA,IAAG,CAAA;AAAA,EAC7F;AACA,EAAA,MAAM,SAASL,KAAAA,GAAO,OAAA,KAAYK,IAAAA,GAAM,IAAA,CAAK,KAAK,CAAC,CAAA,CAAA;AAEnD,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,MAAA,GAAS,KAAK,CAAA,GAAI,WAAA,CAAY,KAAK,GAAA,CAAI,KAAK,GAAG,EAAE,CAAA,CAAA;AACvD,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AASO,SAAS,SAAA,CAAU,CAAA,EAAW,CAAA,EAAW,QAAA,GAAW,IAAA,EAAkB;AAC3E,EAAA,MAAM,EAAA,GAAK,qBAAqB,CAAC,CAAA;AACjC,EAAA,MAAM,EAAA,GAAK,qBAAqB,CAAC,CAAA;AACjC,EAAA,MAAM,KAAK,EAAA,CAAG,MAAA;AACd,EAAA,MAAM,KAAK,EAAA,CAAG,MAAA;AACd,EAAA,IAAI,EAAA,GAAK,CAAA,IAAK,EAAA,GAAK,CAAA,EAAG;AACpB,IAAA,MAAM,IAAI,qBAAA,CAAsB,uDAAA,EAAyD,GAAA,EAAK;AAAA,MAC5F,EAAA;AAAA,MACA;AAAA,KACD,CAAA;AAAA,EACH;AAEA,EAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,EAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,EAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,EAAA,EAAI,KAAK,KAAA,IAAS,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AAC/C,EAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,EAAA,EAAI,KAAK,KAAA,IAAS,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AAC/C,EAAA,KAAA,IAAS,EAAA;AACT,EAAA,KAAA,IAAS,EAAA;AAET,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,EAAA,EAAI,CAAA,EAAA,EAAK;AAC3B,IAAA,MAAM,CAAA,GAAA,CAAK,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACzB,IAAA,GAAA,IAAO,CAAA,GAAI,CAAA;AAAA,EACb;AACA,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,EAAA,EAAI,CAAA,EAAA,EAAK;AAC3B,IAAA,MAAM,CAAA,GAAA,CAAK,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,IAAK,KAAA;AACzB,IAAA,GAAA,IAAO,CAAA,GAAI,CAAA;AAAA,EACb;AACA,EAAA,MAAM,IAAA,GAAO,OAAO,EAAA,GAAK,CAAA,CAAA;AACzB,EAAA,MAAM,IAAA,GAAO,OAAO,EAAA,GAAK,CAAA,CAAA;AAEzB,EAAA,IAAI,KAAA;AACJ,EAAA,IAAI,EAAA;AAEJ,EAAA,IAAI,QAAA,EAAU;AACZ,IAAA,MAAM,SAAA,GAAA,CAAA,CAAc,KAAK,CAAA,IAAK,IAAA,GAAA,CAAQ,KAAK,CAAA,IAAK,IAAA,KAAS,KAAK,EAAA,GAAK,CAAA,CAAA;AACnE,IAAA,MAAM,QAAQ,IAAA,CAAK,IAAA,CAAK,aAAa,CAAA,GAAI,EAAA,GAAK,IAAI,EAAA,CAAG,CAAA;AACrD,IAAA,IAAI,KAAA,KAAU,CAAA;AACZ,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,6CAAA;AAAA,QACA,OAAA;AAAA,QACA;AAAA,OACF;AACF,IAAA,KAAA,GAAA,CAAS,QAAQ,KAAA,IAAS,KAAA;AAC1B,IAAA,EAAA,GAAK,KAAK,EAAA,GAAK,CAAA;AAAA,EACjB,CAAA,MAAO;AACL,IAAA,MAAM,QAAQ,IAAA,CAAK,IAAA,CAAK,IAAA,GAAO,EAAA,GAAK,OAAO,EAAE,CAAA;AAC7C,IAAA,IAAI,KAAA,KAAU,CAAA;AACZ,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,6CAAA;AAAA,QACA,OAAA;AAAA,QACA;AAAA,OACF;AACF,IAAA,KAAA,GAAA,CAAS,QAAQ,KAAA,IAAS,KAAA;AAC1B,IAAA,EAAA,GAAA,CAAM,IAAA,GAAO,EAAA,GAAK,IAAA,GAAO,EAAA,KAAO,CAAA,IAAA,CAAM,IAAA,GAAO,EAAA,KAAO,CAAA,IAAK,EAAA,GAAK,CAAA,CAAA,GAAA,CAAM,IAAA,GAAO,EAAA,KAAO,KAAK,EAAA,GAAK,CAAA,CAAA,CAAA;AAAA,EAC9F;AAEA,EAAA,MAAM,MAAA,GAAS,KAAK,CAAA,GAAI,WAAA,CAAY,KAAK,GAAA,CAAI,KAAK,GAAG,EAAE,CAAA,CAAA;AACvD,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AASO,SAAS,SAAA,CAAU,GAAW,CAAA,EAAuB;AAC1D,EAAA,IAAI,CAAA,CAAE,IAAA,KAAS,CAAA,CAAE,IAAA,EAAM;AACrB,IAAA,MAAM,IAAI,qBAAA,CAAsB,qDAAA,EAAuD,MAAA,EAAQ;AAAA,MAC7F,GAAG,CAAA,CAAE,IAAA;AAAA,MACL,GAAG,CAAA,CAAE;AAAA,KACN,CAAA;AAAA,EACH;AACA,EAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,gDAAA,EAAkD,GAAA,EAAK,CAAC,CAAA;AAAA,EAC1F;AAGA,EAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,CAAC,CAAA;AAChC,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,IAAA,KAAS;AAG9B,IAAA,KAAA,CAAM,CAAC,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA;AAC9B,IAAA,CAAA,EAAA;AAAA,EACF,CAAC,CAAA;AACD,EAAA,MAAM,EAAA,GAAK,IAAI,YAAA,CAAa,CAAC,CAAA;AAC7B,EAAA,CAAA,GAAI,CAAA;AACJ,EAAA,kBAAA,CAAmB,CAAA,EAAG,CAAC,IAAA,KAAS;AAC9B,IAAA,EAAA,CAAG,CAAC,CAAA,GAAI,WAAA,CAAY,CAAA,EAAG,IAAI,CAAA;AAC3B,IAAA,CAAA,EAAA;AAAA,EACF,CAAC,CAAA;AACD,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,KAAA,CAAM,CAAC,KAAK,KAAA,CAAM,CAAC,KAAK,CAAA,KAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,CAAA;AAAA,EACzC;AAEA,EAAA,IAAIL,KAAAA,GAAO,CAAA;AACX,EAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,EAAG,KAAKA,KAAAA,IAAQ,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAChD,EAAAA,KAAAA,IAAQ,CAAA;AAER,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,CAAA,GAAA,CAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,IAAKA,KAAAA;AAC5B,IAAA,EAAA,IAAM,CAAA,GAAI,CAAA;AAAA,EACZ;AACA,EAAA,MAAM,OAAA,GAAU,MAAM,CAAA,GAAI,CAAA,CAAA;AAC1B,EAAA,MAAM,OAAA,GAAU,IAAA,CAAK,IAAA,CAAK,OAAO,CAAA;AACjC,EAAA,IAAI,YAAY,CAAA,EAAG;AACjB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,mDAAA;AAAA,MACA,SAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,MAAM,KAAA,GAAQA,KAAAA,IAAQ,OAAA,GAAU,IAAA,CAAK,KAAK,CAAC,CAAA,CAAA;AAC3C,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,MAAA,GAAS,KAAK,CAAA,GAAI,WAAA,CAAY,KAAK,GAAA,CAAI,KAAK,GAAG,EAAE,CAAA,CAAA;AACvD,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AAUO,SAAS,SAAA,CAAU,OAAe,KAAA,EAA4B;AACnE,EAAA,MAAM,GAAA,GAAM,eAAe,KAAK,CAAA;AAChC,EAAA,MAAM,IAAI,GAAA,CAAI,MAAA;AACd,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,kDAAA,EAAoD,GAAA,EAAK,CAAC,CAAA;AAAA,EAC5F;AACA,EAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,EAAA,IAAI,MAAA,GAAS,CAAA;AAEb,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,CAAA,GAAI,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACpB,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,IAAI,CAAA,EAAG;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,0DAAA;AAAA,QACA,OAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,MAAA,IAAU,CAAA;AAAA,EACZ;AAEA,EAAA,IAAI,KAAA,IAAS,KAAA,CAAM,IAAA,KAAS,KAAA,CAAM,IAAA,EAAM;AACtC,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,kEAAA;AAAA,MACA,MAAA;AAAA,MACA,EAAE,KAAA,EAAO,KAAA,CAAM,IAAA,EAAM,KAAA,EAAO,MAAM,IAAA;AAAK,KACzC;AAAA,EACF;AAEA,EAAA,IAAI,CAAC,KAAA,EAAO;AAEV,IAAA,MAAM,WAAW,MAAA,GAAS,CAAA;AAC1B,IAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,QAAQ,CAAA,IAAK,YAAY,CAAA,EAAG;AAC/C,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,yDAAA;AAAA,QACA,UAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AACA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,CAAA,GAAI,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACpB,MAAA,KAAA,IAAA,CAAU,CAAA,GAAI,aAAa,CAAA,GAAI,QAAA;AAAA,IACjC;AAAA,EACF,CAAA,MAAO;AACL,IAAA,MAAM,GAAA,GAAM,eAAe,KAAK,CAAA;AAChC,IAAA,IAAI,MAAA,GAAS,CAAA;AACb,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,CAAA,GAAI,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACpB,MAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAAS,CAAC,CAAA,IAAK,KAAK,CAAA,EAAG;AACjC,QAAA,MAAM,IAAI,qBAAA;AAAA,UACR,yDAAA;AAAA,UACA,OAAA;AAAA,UACA;AAAA,SACF;AAAA,MACF;AACA,MAAA,MAAA,IAAU,CAAA;AAAA,IACZ;AAEA,IAAA,MAAM,IAAA,GAAO,IAAA,CAAK,IAAA,CAAK,MAAA,CAAO,OAAO,CAAA;AACrC,IAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,GAAA,CAAI,IAAA,CAAK,GAAA,CAAI,MAAM,CAAA,EAAG,IAAA,CAAK,GAAA,CAAI,MAAM,CAAC,CAAA;AACzD,IAAA,IAAI,KAAK,GAAA,CAAI,MAAA,GAAS,MAAM,CAAA,GAAI,OAAO,KAAA,EAAO;AAC5C,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,0EAAA;AAAA,QACA,KAAA;AAAA,QACA,EAAE,KAAA,EAAO,MAAA,EAAQ,KAAA,EAAO,MAAA;AAAO,OACjC;AAAA,IACF;AAEA,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,IAAA,GAAO,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACvB,MAAA,MAAM,IAAA,GAAO,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACvB,MAAA,KAAA,IAAA,CAAU,IAAA,GAAO,SAAS,CAAA,GAAI,IAAA;AAAA,IAChC;AAAA,EACF;AAEA,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,IAAI,KAAK,CAAA,EAAG;AACV,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,8DAAA;AAAA,MACA,IAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,YAAA,CAAa,KAAA,EAAO,EAAE,CAAA;AAEzC,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AAOO,SAAS,MAAA,CAAO,MAAc,GAAA,EAAmD;AACtF,EAAA,MAAM,CAAA,GAAI,qBAAqB,IAAI,CAAA;AACnC,EAAA,MAAM,IAAI,CAAA,CAAE,MAAA;AACZ,EAAA,IAAI,MAAM,CAAA,EAAG;AACX,IAAA,MAAM,IAAI,qBAAA,CAAsB,wCAAA,EAA0C,GAAA,EAAK,CAAC,CAAA;AAAA,EAClF;AAEA,EAAA,IAAI,OAAO,GAAA,KAAQ,QAAA,IAAY,GAAA,KAAQ,MAAA,EAAQ;AAC7C,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,8BAA8B,GAAG,CAAA,kCAAA,CAAA;AAAA,MACjC,KAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,MAAM,CAAA,GAAI,OAAO,GAAA,KAAQ,QAAA,GAAW,CAAC,CAAA,KAAc,SAAA,CAAU,CAAC,CAAA,GAAI,GAAA;AAElE,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,EAAA,GAAK,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA;AACnB,IAAA,MAAM,EAAA,GAAK,EAAE,EAAE,CAAA;AACf,IAAA,MAAM,KAAA,GAAA,CAAS,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,EAAA;AAC5B,IAAA,MAAM,MAAA,GAAS,KAAK,CAAA,GAAI,CAAA;AACxB,IAAA,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAA,EAAG,KAAA,EAAO,MAAM,CAAA;AAAA,EAC/B;AAIA,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,GAAA,EAAK,CAAA,EAAA,EAAK;AAC5B,IAAA,MAAM,IAAA,GAAO,KAAK,GAAA,CAAI,EAAA,GAAK,IAAI,CAAA,GAAI,CAAA,GAAI,IAAI,CAAC,CAAA;AAC5C,IAAA,CAAA,IAAA,CAAM,CAAA,GAAI,CAAA,KAAM,CAAA,GAAI,CAAA,GAAI,EAAA,IAAM,IAAA;AAC9B,IAAA,IAAI,OAAO,KAAA,EAAO;AAAA,EACpB;AACA,EAAA,CAAA,GAAI,IAAA,CAAK,IAAI,CAAA,EAAG,IAAA,CAAK,IAAI,CAAA,EAAG,CAAA,GAAI,CAAC,CAAC,CAAA;AAElC,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,CAAA,EAAE;AACnC;AASO,SAAS,WAAW,CAAA,EAAuB;AAChD,EAAA,MAAM,CAAA,GAAI,qBAAqB,CAAC,CAAA;AAChC,EAAA,MAAM,IAAI,CAAA,CAAE,MAAA;AACZ,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,CAAC,CAAA;AAAA,EACpF;AACA,EAAA,MAAM,EAAE,IAAA,EAAAA,KAAAA,EAAM,EAAA,EAAG,GAAI,UAAU,CAAC,CAAA;AAChC,EAAA,MAAM,MAAM,EAAA,GAAK,CAAA;AACjB,EAAA,MAAMK,IAAAA,GAAM,IAAA,CAAK,IAAA,CAAK,GAAG,CAAA;AACzB,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAASA,IAAG,CAAA,IAAKA,SAAQ,CAAA,EAAG;AACtC,IAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,KAAA,EAAOA,IAAG,CAAA;AAAA,EAC5F;AAEA,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,CAAA,GAAA,CAAK,CAAA,CAAE,CAAC,CAAA,IAAK,CAAA,IAAKL,KAAAA;AACxB,IAAA,EAAA,IAAM,IAAI,CAAA,GAAI,CAAA;AACd,IAAA,EAAA,IAAM,CAAA,GAAI,IAAI,CAAA,GAAI,CAAA;AAAA,EACpB;AACA,EAAA,MAAM,IAAA,GAAO,EAAA,GAAK,CAAA,GAAI,GAAA,IAAO,GAAA;AAC7B,EAAA,MAAM,IAAA,GAAO,EAAA,GAAK,CAAA,IAAK,GAAA,GAAM,GAAA,CAAA;AAG7B,EAAA,MAAM,CAAA,GAAI,IAAA,GAAO,IAAA,CAAK,IAAA,CAAA,CAAO,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,CAAA,IAAO,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,CAAG,CAAA;AAC9D,EAAA,MAAM,QACH,CAAA,IAAK,CAAA,GAAI,CAAA,GAAI,EAAA,GAAK,IAAI,EAAA,CAAA,IAAO,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,OAAQ,CAAA,GAAI,CAAA,KAAM,IAAI,CAAA,CAAA,IAAM,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,CAAA,CAAA;AACxF,EAAA,MAAM,KAAK,EAAA,GAAK,IAAA,CAAK,IAAA,CAAK,CAAA,IAAK,QAAQ,CAAA,CAAE,CAAA;AACzC,EAAA,MAAM,KAAA,GAAQ,IAAI,IAAA,CAAK,IAAA,CAAK,MAAM,IAAA,CAAK,GAAA,CAAI,EAAE,CAAC,CAAA;AAC9C,EAAA,MAAM,KAAA,GAAQ,IAAA,CAAK,IAAA,CAAK,CAAA,IAAK,KAAK,CAAA,CAAE,CAAA;AACpC,EAAA,MAAM,UAAU,CAAA,GAAI,KAAA;AACpB,EAAA,MAAM,EAAA,GAAK,KAAA,GAAQ,IAAA,CAAK,GAAA,CAAI,OAAA,GAAU,KAAK,IAAA,CAAK,OAAA,GAAU,OAAA,GAAU,CAAC,CAAC,CAAA;AAGtE,EAAA,MAAM,CAAA,GAAK,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,IAAO,CAAA,GAAI,CAAA,CAAA;AAC/B,EAAA,MAAM,KAAA,GAAS,EAAA,GAAK,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,CAAA,CAAA,IAAA,CAAQ,CAAA,GAAI,CAAA,KAAM,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,CAAA,CAAA;AACjF,EAAA,MAAM,IAAA,GAAA,CAAQ,IAAA,GAAO,CAAA,IAAK,IAAA,CAAK,KAAK,KAAK,CAAA;AACzC,EAAA,MAAM,SAAA,GACF,KAAK,CAAA,GAAI,CAAA,GAAI,IAAI,CAAA,GAAI,CAAA,CAAA,IAAA,CAAQ,CAAA,GAAI,CAAA,KAAM,CAAA,GAAI,CAAA,CAAA,CAAA,GAC7C,KAAK,IAAA,CAAM,CAAA,IAAK,IAAI,CAAA,CAAA,IAAM,CAAA,GAAI,MAAO,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,IAAM,CAAA,GAAI,CAAA,CAAA,CAAG,CAAA;AAC7D,EAAA,MAAM,KAAA,GAAQ,CAAA,GAAK,CAAA,GAAI,SAAA,IAAc,CAAA,GAAI,SAAA,GAAY,IAAA,CAAK,IAAA,CAAK,CAAA,GAAI,CAAA,IAAK,SAAA,GAAY,SAAA,CAAU,CAAA,CAAA;AAC9F,EAAA,MAAM,KAAA,GAAQ,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,KAAA,CAAA;AAC3B,EAAA,MAAM,QAAQ,CAAA,GAAI,IAAA,GAAO,KAAK,IAAA,CAAK,CAAA,IAAK,QAAQ,CAAA,CAAE,CAAA;AAClD,EAAA,MAAM,QACJ,KAAA,KAAU,CAAA,GAAI,MAAA,CAAO,GAAA,GAAM,KAAK,IAAA,CAAK,KAAK,CAAA,GAAA,CAAA,CAAM,CAAA,GAAI,IAAI,KAAA,IAAS,IAAA,CAAK,GAAA,CAAI,KAAK,OAAO,CAAA,GAAI,CAAA,CAAA;AAC5F,EAAA,MAAM,MAAM,KAAA,GAAQ,KAAA,IAAS,KAAK,IAAA,CAAK,CAAA,IAAK,IAAI,KAAA,CAAM,CAAA;AAEtD,EAAA,MAAM,EAAA,GAAK,EAAA,GAAK,EAAA,GAAK,EAAA,GAAK,EAAA;AAC1B,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,YAAA,CAAa,EAAA,EAAI,CAAC,CAAA;AACrC,EAAA,OAAO,EAAE,SAAA,EAAW,EAAA,EAAI,MAAA,EAAO;AACjC;AAOO,SAAS,QAAQ,CAAA,EAAuB;AAC7C,EAAA,MAAM,MAAA,GAAS,qBAAqB,CAAC,CAAA;AACrC,EAAA,OAAO,YAAY,MAAM,CAAA;AAC3B;AAWO,SAAS,SAAS,CAAA,EAIvB;AACA,EAAA,MAAM,MAAA,GAAS,qBAAqB,CAAC,CAAA;AACrC,EAAA,MAAM,IAAI,MAAA,CAAO,MAAA;AACjB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,0CAAA,EAA4C,GAAA,EAAK,CAAC,CAAA;AAAA,EACpF;AACA,EAAA,MAAM,EAAE,IAAA,EAAAA,KAAAA,EAAM,EAAA,EAAG,GAAI,UAAU,MAAM,CAAA;AACrC,EAAA,MAAMI,SAAAA,GAAW,CAAA,GAAI,CAAA,GAAI,EAAA,IAAM,IAAI,CAAA,CAAA,GAAK,GAAA;AACxC,EAAA,IAAIC,IAAAA,GAAM,IAAA,CAAK,IAAA,CAAKD,SAAQ,CAAA;AAC5B,EAAA,IAAI,CAAC,MAAA,CAAO,QAAA,CAASC,IAAG,CAAA,IAAKA,SAAQ,CAAA,EAAG;AACtC,IAAA,MAAM,IAAI,qBAAA,CAAsB,4CAAA,EAA8C,KAAA,EAAOA,IAAG,CAAA;AAAA,EAC1F;AACA,EAAA,IAAI,IAAI,EAAA,EAAI;AACV,IAAA,MAAMC,SAAAA,GAAW,CAAC,CAAA,KAAsB;AACtC,MAAA,IAAI,CAAA,KAAM,CAAA,EAAG,OAAO,MAAA,CAAO,CAAC,CAAA,IAAK,CAAA;AACjC,MAAA,MAAM,GAAA,GAAA,CAAO,IAAI,CAAA,IAAK,CAAA;AACtB,MAAA,MAAM,EAAA,GAAK,IAAA,CAAK,KAAA,CAAM,GAAG,CAAA;AACzB,MAAA,MAAM,EAAA,GAAK,IAAA,CAAK,IAAA,CAAK,GAAG,CAAA;AACxB,MAAA,MAAM,EAAA,GAAK,MAAA,CAAO,EAAE,CAAA,IAAK,CAAA;AACzB,MAAA,MAAM,EAAA,GAAK,MAAA,CAAO,EAAE,CAAA,IAAK,EAAA;AACzB,MAAA,OAAO,EAAA,GAAA,CAAM,GAAA,GAAM,EAAA,KAAO,EAAA,GAAK,EAAA,CAAA;AAAA,IACjC,CAAA;AACA,IAAA,MAAM,EAAA,GAAKA,UAAS,IAAI,CAAA;AACxB,IAAA,MAAM,EAAA,GAAKA,UAAS,IAAI,CAAA;AACxB,IAAA,MAAM,MAAM,EAAA,GAAK,EAAA;AACjB,IAAA,MAAM,SAAS,GAAA,GAAM,KAAA;AACrB,IAAA,IAAI,MAAA,CAAO,QAAA,CAAS,MAAM,CAAA,IAAK,SAAS,CAAA,EAAG;AACzC,MAAAD,IAAAA,GAAM,MAAA;AAAA,IACR;AAAA,EACF;AAGA,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,EAAA,GAAA,CAAA,CAAO,MAAA,CAAO,CAAC,CAAA,IAAK,KAAKL,KAAAA,IAAQK,IAAAA;AACvC,IAAA,MAAM,OAAO,MAAA,CAAO,CAAA,GAAI,IAAI,CAAC,CAAA,IAAK,KAAKL,KAAAA,IAAQK,IAAAA;AAC/C,IAAA,MAAM,IAAA,GAAO,IAAA,CAAK,GAAA,CAAI,MAAA,EAAQ,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,KAAA,EAAO,SAAA,CAAU,EAAE,CAAC,CAAC,CAAA;AAChE,IAAA,MAAM,IAAA,GAAO,IAAA,CAAK,GAAA,CAAI,MAAA,EAAQ,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,KAAA,EAAO,SAAA,CAAU,EAAE,CAAC,CAAC,CAAA;AAChE,IAAA,EAAA,IAAA,CAAO,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,GAAK,CAAA,KAAM,IAAA,CAAK,GAAA,CAAI,IAAI,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,CAAA,GAAI,IAAI,CAAA,CAAA;AAAA,EAC/D;AACA,EAAA,EAAA,GAAK,CAAC,IAAI,EAAA,GAAK,CAAA;AAEf,EAAA,MAAM,eAAe,CAAC,KAAA,EAAO,KAAA,EAAO,KAAA,EAAO,OAAO,KAAK,CAAA;AACvD,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,CAAA,GAAI,CAAA,GAAI,MAAM,CAAA,GAAI,CAAA,CAAA;AACrC,EAAA,MAAM,eAAA,GAAkB,YAAA,CAAa,GAAA,CAAI,CAAC,CAAA,KAAM,IAAA,CAAK,KAAA,CAAO,CAAA,GAAI,MAAA,GAAU,GAAI,CAAA,GAAI,GAAI,CAAA;AAEtF,EAAA,OAAO;AAAA,IACL,SAAA,EAAW,EAAA;AAAA,IACX,eAAA;AAAA,IACA,oBAAoB,CAAC,IAAA,EAAM,GAAA,EAAK,IAAA,EAAM,OAAO,IAAI;AAAA,GACnD;AACF;AAYO,SAAS,YAAA,CAAa,GAAW,CAAA,EAAuB;AAC7D,EAAA,MAAM,KAAK,CAAA,CAAE,IAAA;AACb,EAAA,MAAM,KAAK,CAAA,CAAE,IAAA;AAEb,EAAA,IAAI,EAAA,GAAK,CAAA,IAAK,EAAA,GAAK,CAAA,EAAG;AACpB,IAAA,MAAM,IAAI,qBAAA,CAAsB,gCAAA,EAAkC,MAAA,EAAQ;AAAA,MACxE,CAAA,EAAG,EAAA;AAAA,MACH,CAAA,EAAG;AAAA,KACJ,CAAA;AAAA,EACH;AAEA,EAAA,MAAM,KAAA,GAAQ,eAAe,CAAC,CAAA;AAC9B,EAAA,MAAM,KAAA,GAAQ,eAAe,CAAC,CAAA;AAC9B,EAAA,MAAM,IAAI,EAAA,GAAK,EAAA;AACf,EAAA,MAAM,QAAA,GAAW,IAAI,YAAA,CAAa,CAAC,CAAA;AACnC,EAAA,QAAA,CAAS,GAAA,CAAI,OAAO,CAAC,CAAA;AACrB,EAAA,QAAA,CAAS,GAAA,CAAI,OAAO,EAAE,CAAA;AAEtB,EAAA,MAAM,EAAE,KAAA,EAAO,MAAA,EAAO,GAAI,SAAS,QAAQ,CAAA;AAG3C,EAAA,IAAI,EAAA,GAAK,CAAA;AACT,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,EAAA,EAAI,CAAA,EAAA,EAAK;AAC3B,IAAA,EAAA,IAAM,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAAA,EACpB;AAEA,EAAA,MAAM,EAAA,GAAK,EAAA,GAAM,EAAA,IAAM,EAAA,GAAK,CAAA,CAAA,GAAM,CAAA;AAClC,EAAA,MAAM,EAAA,GAAK,KAAK,EAAA,GAAK,EAAA;AACrB,EAAA,MAAM,CAAA,GAAI,IAAA,CAAK,GAAA,CAAI,EAAA,EAAI,EAAE,CAAA;AAEzB,EAAA,MAAM,KAAA,GAAS,KAAK,EAAA,GAAM,CAAA;AAE1B,EAAA,MAAM,SAAS,CAAA,GAAI,CAAA,GAAI,MAAA,IAAU,CAAA,IAAK,IAAI,CAAA,CAAA,CAAA,GAAM,CAAA;AAChD,EAAA,MAAM,IAAA,GAAQ,EAAA,GAAK,EAAA,IAAM,CAAA,GAAI,IAAI,MAAA,CAAA,GAAW,EAAA;AAC5C,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAQ,GAAA,EAAI;AAAA,EACrC;AAEA,EAAA,MAAM,IAAA,GAAO,IAAA,CAAK,IAAA,CAAK,IAAI,CAAA;AAG3B,EAAA,MAAM,aAAA,GAAgB,KAAK,EAAA,GAAK,EAAA;AAChC,EAAA,MAAM,UAAA,GAAa,gBAAiB,CAAA,GAAI,KAAA,GAAQ,MAAM,CAAA,GAAI,KAAA,GAAQ,OAAO,CAAA,GAAK,CAAA;AAC9E,EAAA,MAAM,CAAA,GAAA,CAAK,CAAA,GAAI,KAAA,GAAQ,UAAA,IAAc,IAAA;AACrC,EAAA,MAAM,SAAS,CAAA,IAAK,CAAA,GAAI,UAAU,IAAA,CAAK,GAAA,CAAI,CAAC,CAAC,CAAA,CAAA;AAE7C,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAO;AAChC;AAUO,SAAS,QAAA,CAAS,GAAW,CAAA,EAAwB;AAC1D,EAAA,MAAM,IAAI,CAAA,CAAE,IAAA;AACZ,EAAA,MAAM,QAAkB,EAAC;AAEzB,EAAA,IAAI,CAAA,EAAG;AACL,IAAA,IAAI,CAAA,CAAE,IAAA,KAAS,CAAA,CAAE,IAAA,EAAM;AACrB,MAAA,MAAM,IAAI,qBAAA,CAAsB,uCAAA,EAAyC,MAAA,EAAQ;AAAA,QAC/E,GAAG,CAAA,CAAE,IAAA;AAAA,QACL,GAAG,CAAA,CAAE;AAAA,OACN,CAAA;AAAA,IACH;AACA,IAAA,MAAM,EAAA,GAAK,eAAe,CAAC,CAAA;AAC3B,IAAA,MAAM,EAAA,GAAK,eAAe,CAAC,CAAA;AAC3B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,QAAQ,EAAA,CAAG,CAAC,KAAK,CAAA,KAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA,CAAA;AACtC,MAAA,IAAI,IAAA,KAAS,CAAA,EAAG,KAAA,CAAM,IAAA,CAAK,IAAI,CAAA;AAAA,IACjC;AAAA,EACF,CAAA,MAAO;AACL,IAAA,MAAM,EAAA,GAAK,eAAe,CAAC,CAAA;AAC3B,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,GAAA,GAAM,EAAA,CAAG,CAAC,CAAA,IAAK,CAAA;AACrB,MAAA,IAAI,GAAA,KAAQ,CAAA,EAAG,KAAA,CAAM,IAAA,CAAK,GAAG,CAAA;AAAA,IAC/B;AAAA,EACF;AAEA,EAAA,IAAI,KAAA,CAAM,WAAW,CAAA,EAAG;AACtB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,uDAAA;AAAA,MACA,OAAA;AAAA,MACA,KAAA,CAAM;AAAA,KACR;AAAA,EACF;AAEA,EAAA,MAAM,QAAA,GAAW,IAAI,YAAA,CAAa,KAAA,CAAM,MAAM,CAAA;AAC9C,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,IAAA,QAAA,CAAS,CAAC,CAAA,GAAI,IAAA,CAAK,IAAI,KAAA,CAAM,CAAC,KAAK,CAAC,CAAA;AAAA,EACtC;AAEA,EAAA,MAAM,EAAE,KAAA,EAAO,MAAA,EAAO,GAAI,SAAS,QAAQ,CAAA;AAE3C,EAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,KAAA,CAAM,QAAQ,CAAA,EAAA,EAAK;AACrC,IAAA,IAAA,CAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,IAAK,CAAA,EAAG;AACvB,MAAA,KAAA,IAAS,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAAA,IACvB;AAAA,EACF;AAEA,EAAA,MAAM,OAAO,KAAA,CAAM,MAAA;AACnB,EAAA,MAAM,KAAA,GAAS,IAAA,IAAQ,IAAA,GAAO,CAAA,CAAA,GAAM,CAAA;AAEpC,EAAA,MAAM,IAAA,GAAQ,QAAQ,IAAA,GAAO,CAAA,CAAA,IAAM,IAAI,IAAA,GAAO,CAAA,CAAA,GAAM,KAAK,MAAA,GAAS,EAAA;AAClE,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAQ,GAAA,EAAI;AAAA,EACzC;AAEA,EAAA,MAAM,IAAA,GAAO,IAAA,CAAK,IAAA,CAAK,IAAI,CAAA;AAG3B,EAAA,MAAM,gBAAgB,IAAA,GAAO,EAAA;AAC7B,EAAA,MAAM,UAAA,GAAa,gBAAiB,KAAA,GAAQ,KAAA,GAAQ,MAAM,KAAA,GAAQ,KAAA,GAAQ,OAAO,CAAA,GAAK,CAAA;AACtF,EAAA,MAAM,CAAA,GAAA,CAAK,KAAA,GAAQ,KAAA,GAAQ,UAAA,IAAc,IAAA;AACzC,EAAA,MAAM,SAAS,CAAA,IAAK,CAAA,GAAI,UAAU,IAAA,CAAK,GAAA,CAAI,CAAC,CAAC,CAAA,CAAA;AAE7C,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AASO,SAAS,WAAW,OAAA,EAA+B;AACxD,EAAA,MAAM,IAAI,OAAA,CAAQ,MAAA;AAClB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,GAAA,EAAK,CAAC,CAAA;AAAA,EAChF;AAEA,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,MAAM,KAAA,GAAQ,IAAI,KAAA,CAA0B,CAAC,CAAA;AAC7C,EAAA,MAAM,gBAAA,GAAmB,IAAI,KAAA,CAA0B,CAAC,CAAA;AACxD,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,CAAC,MAAA,IAAU,MAAA,CAAO,IAAA,GAAO,CAAA,EAAG;AAC9B,MAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,MAAA,EAAQ;AAAA,QAC9E,KAAA,EAAO,CAAA;AAAA,QACP,IAAA,EAAM,QAAQ,IAAA,IAAQ;AAAA,OACvB,CAAA;AAAA,IACH;AACA,IAAA,gBAAA,CAAiB,CAAC,CAAA,GAAI,MAAA;AACtB,IAAA,KAAA,CAAM,CAAC,IAAI,MAAA,CAAO,IAAA;AAClB,IAAA,CAAA,IAAK,MAAA,CAAO,IAAA;AAAA,EACd;AAEA,EAAA,MAAM,QAAA,GAAW,IAAI,YAAA,CAAa,CAAC,CAAA;AACnC,EAAA,MAAM,UAAA,GAAa,IAAI,UAAA,CAAW,CAAC,CAAA;AACnC,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,iBAAiB,CAAC,CAAA;AACjC,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,qBAAA,CAAsB,sCAAA,EAAwC,QAAA,EAAU,CAAC,CAAA;AAAA,IACrF;AACA,IAAA,MAAM,IAAA,GAAO,eAAe,MAAM,CAAA;AAClC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,IAAA,CAAK,QAAQ,CAAA,EAAA,EAAK;AACpC,MAAA,QAAA,CAAS,GAAG,CAAA,GAAI,IAAA,CAAK,CAAC,CAAA,IAAK,CAAA;AAC3B,MAAA,UAAA,CAAW,GAAG,CAAA,GAAI,CAAA;AAClB,MAAA,GAAA,EAAA;AAAA,IACF;AAAA,EACF;AAEA,EAAA,MAAM,EAAE,KAAA,EAAO,MAAA,EAAO,GAAI,SAAS,QAAQ,CAAA;AAE3C,EAAA,MAAM,QAAA,GAAW,IAAI,YAAA,CAAa,CAAC,CAAA;AACnC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,KAAA,GAAQ,UAAA,CAAW,CAAC,CAAA,IAAK,CAAA;AAC/B,IAAA,MAAM,IAAA,GAAO,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AACzB,IAAA,QAAA,CAAS,KAAK,CAAA,GAAA,CAAK,QAAA,CAAS,KAAK,KAAK,CAAA,IAAK,IAAA;AAAA,EAC7C;AAEA,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,EAAA,GAAK,QAAA,CAAS,CAAC,CAAA,IAAK,CAAA;AAC1B,IAAA,MAAM,EAAA,GAAK,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AACvB,IAAA,CAAA,IAAM,KAAK,EAAA,GAAM,EAAA;AAAA,EACnB;AACA,EAAA,CAAA,GAAK,MAAM,CAAA,IAAK,CAAA,GAAI,CAAA,CAAA,CAAA,GAAO,CAAA,GAAI,KAAK,CAAA,GAAI,CAAA,CAAA;AAGxC,EAAA,MAAM,aAAA,GAAgB,IAAI,CAAA,GAAI,CAAA,GAAI,UAAU,CAAA,GAAI,CAAA,GAAI,IAAI,CAAA,CAAA,GAAK,CAAA;AAC7D,EAAA,IAAI,iBAAiB,CAAA,EAAG;AACtB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,uDAAA;AAAA,MACA,eAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,CAAA,IAAK,aAAA;AAEL,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,YAAA,CAAa,CAAA,EAAG,EAAE,CAAA;AAErC,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAO;AAChC;AASO,SAAS,qBAAqB,OAAA,EAA+B;AAClE,EAAA,MAAM,IAAI,OAAA,CAAQ,MAAA;AAClB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,yDAAA;AAAA,MACA,GAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,MAAM,CAAA,GAAI,OAAA,CAAQ,CAAC,CAAA,EAAG,IAAA,IAAQ,CAAA;AAC9B,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,0DAAA;AAAA,MACA,GAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AAEA,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,MAAA,IAAU,MAAA,CAAO,IAAA,KAAS,CAAA,EAAG;AAC/B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,yDAAA;AAAA,QACA,MAAA;AAAA,QACA,EAAE,QAAA,EAAU,CAAA,EAAG,KAAK,MAAA,CAAO,IAAA,EAAM,aAAa,CAAA;AAAE,OAClD;AAAA,IACF;AAAA,EACF;AAGA,EAAA,MAAM,eAAe,OAAA,CAAQ,GAAA;AAAA,IAAI,CAAC,WAChC,MAAA,GAAS,cAAA,CAAe,MAAM,CAAA,GAAI,IAAI,aAAa,CAAC;AAAA,GACtD;AACA,EAAA,MAAM,QAAA,GAAW,IAAI,YAAA,CAAa,CAAC,CAAA;AACnC,EAAA,IAAI,MAAA,GAAS,CAAA;AAEb,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,KAAA,GAAQ,IAAI,YAAA,CAAa,CAAC,CAAA;AAChC,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,GAAA,GAAM,aAAa,CAAC,CAAA;AAC1B,MAAA,KAAA,CAAM,CAAC,CAAA,GAAI,GAAA,GAAM,CAAC,CAAA,IAAK,CAAA;AAAA,IACzB;AACA,IAAA,MAAM,MAAA,GAAS,SAAS,KAAK,CAAA;AAC7B,IAAA,MAAA,IAAU,MAAA,CAAO,MAAA;AACjB,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,IAAA,GAAO,MAAA,CAAO,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA;AAChC,MAAA,QAAA,CAAS,CAAC,CAAA,GAAA,CAAK,QAAA,CAAS,CAAC,KAAK,CAAA,IAAK,IAAA;AAAA,IACrC;AAAA,EACF;AAEA,EAAA,IAAI,KAAA,GAAQ,CAAA;AACZ,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,EAAA,GAAK,QAAA,CAAS,CAAC,CAAA,IAAK,CAAA;AAC1B,IAAA,KAAA,IAAS,EAAA,GAAK,EAAA;AAAA,EAChB;AACA,EAAA,KAAA,GAAS,EAAA,IAAM,IAAI,CAAA,IAAK,CAAA,GAAI,MAAO,KAAA,GAAQ,CAAA,GAAI,KAAK,CAAA,GAAI,CAAA,CAAA;AAGxD,EAAA,MAAM,aAAA,GAAgB,IAAI,CAAA,GAAI,CAAA,GAAI,UAAU,CAAA,GAAI,CAAA,IAAK,CAAA,GAAI,CAAA,GAAI,CAAA,CAAA,CAAA,GAAM,CAAA;AACnE,EAAA,IAAI,iBAAiB,CAAA,EAAG;AACtB,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,+EAAA;AAAA,MACA,eAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,KAAA,IAAS,aAAA;AAET,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,YAAA,CAAa,KAAA,EAAO,EAAE,CAAA;AAEzC,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AAyBO,SAAS,MAAA,CAAO,WAA0C,OAAA,EAA+B;AAC9F,EAAA,MAAM,IAAI,OAAA,CAAQ,MAAA;AAClB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,qCAAA,EAAuC,GAAA,EAAK,CAAC,CAAA;AAAA,EAC/E;AAGA,EAAA,MAAM,SAAyB,EAAC;AAChC,EAAA,MAAM,UAAoB,EAAC;AAE3B,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,CAAC,MAAA,IAAU,MAAA,CAAO,IAAA,KAAS,CAAA,EAAG;AAChC,MAAA,MAAM,IAAI,qBAAA,CAAsB,8CAAA,EAAgD,WAAA,EAAa;AAAA,QAC3F,KAAA,EAAO,CAAA;AAAA,QACP,IAAA,EAAM,QAAQ,IAAA,IAAQ;AAAA,OACvB,CAAA;AAAA,IACH;AACA,IAAA,MAAM,GAAA,GAAM,qBAAqB,MAAM,CAAA;AACvC,IAAA,IAAI,GAAA,CAAI,SAAS,CAAA,EAAG;AAClB,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,WAAA;AAAA,QACA,GAAA,CAAI;AAAA,OACN;AAAA,IACF;AACA,IAAA,MAAA,CAAO,KAAK,GAAG,CAAA;AAGf,IAAA,IAAI,WAAW,MAAA,EAAQ;AACrB,MAAA,IAAI,GAAA,GAAM,CAAA;AACV,MAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,GAAA,CAAI,QAAQ,CAAA,EAAA,EAAK,GAAA,IAAO,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AACtD,MAAA,OAAA,CAAQ,IAAA,CAAK,GAAA,GAAM,GAAA,CAAI,MAAM,CAAA;AAAA,IAC/B,CAAA,MAAA,IAAW,WAAW,QAAA,EAAU;AAC9B,MAAA,MAAM,GAAA,GAAM,IAAA,CAAK,KAAA,CAAM,GAAA,CAAI,SAAS,CAAC,CAAA;AACrC,MAAA,IAAI,GAAA,CAAI,MAAA,GAAS,CAAA,KAAM,CAAA,EAAG;AACxB,QAAA,OAAA,CAAQ,IAAA,CAAA,CAAA,CAAO,GAAA,CAAI,GAAA,GAAM,CAAC,CAAA,IAAK,MAAM,GAAA,CAAI,GAAG,CAAA,IAAK,CAAA,CAAA,IAAM,CAAC,CAAA;AAAA,MAC1D,CAAA,MAAO;AACL,QAAA,OAAA,CAAQ,IAAA,CAAK,GAAA,CAAI,GAAG,CAAA,IAAK,CAAC,CAAA;AAAA,MAC5B;AAAA,IACF,CAAA,MAAO;AAEL,MAAA,MAAM,SAAA,GAAY,IAAA,CAAK,KAAA,CAAM,GAAA,CAAI,SAAS,GAAG,CAAA;AAC7C,MAAA,IAAI,GAAA,GAAM,CAAA;AACV,MAAA,MAAM,CAAA,GAAI,GAAA,CAAI,MAAA,GAAS,CAAA,GAAI,SAAA;AAC3B,MAAA,KAAA,IAAS,IAAI,SAAA,EAAW,CAAA,GAAI,GAAA,CAAI,MAAA,GAAS,WAAW,CAAA,EAAA,EAAK;AACvD,QAAA,GAAA,IAAO,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AAAA,MACnB;AACA,MAAA,OAAA,CAAQ,IAAA,CAAK,MAAM,CAAC,CAAA;AAAA,IACtB;AAAA,EACF;AAGA,EAAA,MAAM,IAAoB,EAAC;AAC3B,EAAA,MAAM,cAAwB,EAAC;AAC/B,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,IAAI,SAAA,GAAY,CAAA;AAEhB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,IAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,IAAA,IAAI,CAAC,GAAA,EAAK;AACV,IAAA,MAAM,CAAA,GAAI,OAAA,CAAQ,CAAC,CAAA,IAAK,CAAA;AACxB,IAAA,MAAM,IAAA,GAAO,IAAI,YAAA,CAAa,GAAA,CAAI,MAAM,CAAA;AACxC,IAAA,IAAI,IAAA,GAAO,CAAA;AAEX,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,GAAA,CAAI,QAAQ,CAAA,EAAA,EAAK;AACnC,MAAA,MAAM,SAAS,IAAA,CAAK,GAAA,CAAA,CAAK,IAAI,CAAC,CAAA,IAAK,KAAK,CAAC,CAAA;AACzC,MAAA,IAAA,CAAK,CAAC,CAAA,GAAI,MAAA;AACV,MAAA,IAAA,IAAQ,MAAA;AAAA,IACV;AAEA,IAAA,CAAA,CAAE,KAAK,IAAI,CAAA;AACX,IAAA,WAAA,CAAY,IAAA,CAAK,IAAA,GAAO,GAAA,CAAI,MAAM,CAAA;AAClC,IAAA,CAAA,IAAK,GAAA,CAAI,MAAA;AACT,IAAA,SAAA,IAAa,IAAA;AAAA,EACf;AAEA,EAAA,MAAM,aAAa,SAAA,GAAY,CAAA;AAG/B,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,GAAA,GAAM,CAAA;AAEV,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,CAAE,QAAQ,CAAA,EAAA,EAAK;AACjC,IAAA,MAAM,IAAA,GAAO,EAAE,CAAC,CAAA;AAChB,IAAA,IAAI,CAAC,IAAA,EAAM;AACX,IAAA,MAAM,IAAI,IAAA,CAAK,MAAA;AACf,IAAA,GAAA,IAAO,CAAA,GAAA,CAAA,CAAM,WAAA,CAAY,CAAC,CAAA,IAAK,KAAK,UAAA,KAAe,CAAA;AAEnD,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,GAAA,IAAA,CAAA,CAAS,KAAK,CAAC,CAAA,IAAK,MAAM,WAAA,CAAY,CAAC,KAAK,CAAA,CAAA,KAAO,CAAA;AAAA,IACrD;AAAA,EACF;AAEA,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,IAAI,OAAO,CAAA,EAAG;AACZ,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,uDAAA;AAAA,MACA,KAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,IAAI,QAAQ,CAAA,EAAG;AACb,IAAA,OAAO,EAAE,SAAA,EAAW,QAAA,EAAU,MAAA,EAAQ,CAAA,EAAE;AAAA,EAC1C;AACA,EAAA,MAAM,CAAA,GAAI,GAAA,GAAM,GAAA,IAAO,GAAA,GAAM,GAAA,CAAA;AAE7B,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,IAAA,CAAK,CAAA,EAAG,KAAK,GAAG,CAAA;AAEnC,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAO;AAChC;AAuBO,SAAS,YAAY,OAAA,EAA+B;AACzD,EAAA,MAAM,IAAI,OAAA,CAAQ,MAAA;AAClB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,uCAAA,EAAyC,GAAA,EAAK,CAAC,CAAA;AAAA,EACjF;AAGA,EAAA,MAAM,YAAsB,EAAC;AAC7B,EAAA,MAAM,QAAkB,EAAC;AACzB,EAAA,IAAI,CAAA,GAAI,CAAA;AAER,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,CAAC,MAAA,IAAU,MAAA,CAAO,IAAA,KAAS,CAAA,EAAG;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,WAAA;AAAA,QACA,EAAE,KAAA,EAAO,CAAA,EAAG,IAAA,EAAM,MAAA,EAAQ,QAAQ,CAAA;AAAE,OACtC;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,qBAAqB,MAAM,CAAA;AACvC,IAAA,MAAM,IAAI,GAAA,CAAI,MAAA;AACd,IAAA,IAAI,IAAI,CAAA,EAAG;AACT,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,kDAAA;AAAA,QACA,WAAA;AAAA,QACA;AAAA,OACF;AAAA,IACF;AAGA,IAAA,IAAIL,KAAAA,GAAO,CAAA;AACX,IAAA,KAAA,IAAS,CAAA,GAAI,GAAG,CAAA,GAAI,CAAA,EAAG,KAAKA,KAAAA,IAAQ,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AAC9C,IAAAA,KAAAA,IAAQ,CAAA;AAGR,IAAA,IAAI,EAAA,GAAK,CAAA;AACT,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,MAAM,CAAA,GAAA,CAAK,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA,IAAKA,KAAAA;AAC1B,MAAA,EAAA,IAAM,CAAA,GAAI,CAAA;AAAA,IACZ;AACA,IAAA,MAAMI,SAAAA,GAAW,MAAM,CAAA,GAAI,CAAA,CAAA;AAE3B,IAAA,SAAA,CAAU,KAAKA,SAAQ,CAAA;AACvB,IAAA,KAAA,CAAM,KAAK,CAAC,CAAA;AACZ,IAAA,CAAA,IAAK,CAAA;AAAA,EACP;AAGA,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,IAAA,CAAK,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA,MAAO,CAAA,EAAG;AAC7B,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,wDAAA;AAAA,QACA,UAAA;AAAA,QACA,UAAU,CAAC;AAAA,OACb;AAAA,IACF;AAAA,EACF;AAGA,EAAA,IAAI,eAAA,GAAkB,CAAA;AACtB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,eAAA,IAAA,CAAA,CAAqB,MAAM,CAAC,CAAA,IAAK,KAAK,CAAA,KAAM,SAAA,CAAU,CAAC,CAAA,IAAK,CAAA,CAAA;AAAA,EAC9D;AACA,EAAA,MAAM,cAAA,GAAiB,mBAAmB,CAAA,GAAI,CAAA,CAAA;AAI9C,EAAA,IAAI,SAAA,GAAY,CAAA;AAChB,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,SAAA,IAAA,CAAA,CAAe,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,IAAK,CAAA,IAAK,KAAK,GAAA,CAAI,SAAA,CAAU,CAAC,CAAA,IAAK,CAAC,CAAA;AAAA,EACjE;AACA,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA,IAAK,IAAA,CAAK,GAAA,CAAI,cAAc,CAAA,GAAI,SAAA;AAG/C,EAAA,IAAI,QAAA,GAAW,CAAA;AACf,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,QAAA,IAAY,CAAA,IAAA,CAAM,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,IAAK,CAAA,CAAA;AAAA,EACrC;AACA,EAAA,MAAM,CAAA,GAAI,IAAK,CAAA,IAAK,CAAA,IAAK,IAAI,CAAA,CAAA,CAAA,IAAQ,QAAA,GAAW,KAAK,CAAA,GAAI,CAAA,CAAA,CAAA;AAGzD,EAAA,MAAM,QAAQ,CAAA,GAAI,CAAA;AAElB,EAAA,MAAM,KAAK,CAAA,GAAI,CAAA;AACf,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,YAAA,CAAa,KAAA,EAAO,EAAE,CAAA;AAEzC,EAAA,OAAO,EAAE,SAAA,EAAW,KAAA,EAAO,MAAA,EAAO;AACpC;AAOO,SAAS,YAAY,OAAA,EAA+B;AACzD,EAAA,MAAM,IAAI,OAAA,CAAQ,MAAA;AAClB,EAAA,IAAI,IAAI,CAAA,EAAG;AACT,IAAA,MAAM,IAAI,qBAAA,CAAsB,uCAAA,EAAyC,QAAA,EAAU,CAAC,CAAA;AAAA,EACtF;AAEA,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,CAAC,MAAA,IAAU,MAAA,CAAO,IAAA,KAAS,CAAA,EAAG;AAChC,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,WAAA;AAAA,QACA,EAAE,KAAA,EAAO,CAAA,EAAG,IAAA,EAAM,MAAA,EAAQ,QAAQ,CAAA;AAAE,OACtC;AAAA,IACF;AAAA,EACF;AAEA,EAAA,IAAI,CAAA,GAAI,CAAA;AACR,EAAA,MAAM,QAAkB,EAAC;AACzB,EAAA,MAAM,QAAkB,EAAC;AACzB,EAAA,MAAM,SAAyB,EAAC;AAGhC,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,MAAM,MAAA,GAAS,QAAQ,CAAC,CAAA;AACxB,IAAA,IAAI,CAAC,MAAA,EAAQ;AACX,MAAA,MAAM,IAAI,qBAAA;AAAA,QACR,gDAAA;AAAA,QACA,WAAA;AAAA,QACA,EAAE,KAAA,EAAO,CAAA,EAAG,IAAA,EAAM,CAAA;AAAE,OACtB;AAAA,IACF;AACA,IAAA,MAAM,GAAA,GAAM,eAAe,MAAM,CAAA;AACjC,IAAA,MAAM,IAAI,GAAA,CAAI,MAAA;AACd,IAAA,MAAA,CAAO,KAAK,GAAG,CAAA;AACf,IAAA,KAAA,CAAM,KAAK,CAAC,CAAA;AACZ,IAAA,CAAA,IAAK,CAAA;AAEL,IAAA,IAAI,GAAA,GAAM,CAAA;AACV,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,GAAA,IAAO,GAAA,CAAI,CAAC,CAAA,IAAK,CAAA;AAAA,IACnB;AACA,IAAA,KAAA,CAAM,IAAA,CAAK,MAAM,CAAC,CAAA;AAAA,EACpB;AAGA,EAAA,IAAI,QAAA,GAAW,CAAA;AACf,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,IAAA,QAAA,IAAA,CAAa,MAAM,CAAC,CAAA,IAAK,CAAA,KAAM,KAAA,CAAM,CAAC,CAAA,IAAK,CAAA,CAAA;AAAA,EAC7C;AACA,EAAA,MAAM,YAAY,QAAA,GAAW,CAAA;AAG7B,EAAA,IAAI,GAAA,GAAM,CAAA;AACV,EAAA,IAAI,GAAA,GAAM,CAAA;AAEV,EAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,MAAA,CAAO,QAAQ,CAAA,EAAA,EAAK;AACtC,IAAA,MAAM,GAAA,GAAM,OAAO,CAAC,CAAA;AACpB,IAAA,IAAI,CAAC,GAAA,EAAK;AACV,IAAA,MAAM,IAAI,GAAA,CAAI,MAAA;AACd,IAAA,GAAA,IAAO,CAAA,GAAA,CAAA,CAAM,KAAA,CAAM,CAAC,CAAA,IAAK,KAAK,SAAA,KAAc,CAAA;AAE5C,IAAA,KAAA,IAAS,CAAA,GAAI,CAAA,EAAG,CAAA,GAAI,CAAA,EAAG,CAAA,EAAA,EAAK;AAC1B,MAAA,GAAA,IAAA,CAAA,CAAS,IAAI,CAAC,CAAA,IAAK,MAAM,KAAA,CAAM,CAAC,KAAK,CAAA,CAAA,KAAO,CAAA;AAAA,IAC9C;AAAA,EACF;AAEA,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,MAAM,MAAM,CAAA,GAAI,CAAA;AAChB,EAAA,IAAI,OAAO,CAAA,EAAG;AACZ,IAAA,MAAM,IAAI,qBAAA;AAAA,MACR,kEAAA;AAAA,MACA,KAAA;AAAA,MACA;AAAA,KACF;AAAA,EACF;AACA,EAAA,MAAM,MAAM,GAAA,GAAM,GAAA;AAClB,EAAA,MAAM,MAAM,GAAA,GAAM,GAAA;AAClB,EAAA,IAAI,QAAQ,CAAA,EAAG;AAEb,IAAA,MAAMG,EAAAA,GAAI,GAAA,KAAQ,CAAA,GAAI,GAAA,GAAM,QAAA;AAC5B,IAAA,OAAO,EAAE,SAAA,EAAWA,EAAAA,EAAG,QAAQ,GAAA,KAAQ,CAAA,GAAI,MAAM,CAAA,EAAE;AAAA,EACrD;AACA,EAAA,MAAM,IAAI,GAAA,GAAM,GAAA;AAEhB,EAAA,MAAM,MAAA,GAAS,CAAA,GAAI,IAAA,CAAK,CAAA,EAAG,KAAK,GAAG,CAAA;AAEnC,EAAA,OAAO,EAAE,SAAA,EAAW,CAAA,EAAG,MAAA,EAAO;AAChC","file":"chunk-XMWVME2W.js","sourcesContent":["// Correlation\nexport { corrcoef, cov, kendalltau, pearsonr, spearmanr } from \"./correlation\";\n\n// Descriptive\nexport {\n  geometricMean,\n  harmonicMean,\n  kurtosis,\n  mean,\n  median,\n  mode,\n  moment,\n  percentile,\n  quantile,\n  skewness,\n  std,\n  trimMean,\n  variance,\n} from \"./descriptive\";\n\n// Tests\nexport type { TestResult } from \"./tests\";\nexport {\n  anderson,\n  bartlett,\n  chisquare,\n  f_oneway,\n  friedmanchisquare,\n  kruskal,\n  kstest,\n  levene,\n  mannwhitneyu,\n  normaltest,\n  shapiro,\n  ttest_1samp,\n  ttest_ind,\n  ttest_rel,\n  wilcoxon,\n} from \"./tests\";\n","/**\n * Internal utilities for the stats package.\n *\n * This module contains internal helper functions used by the stats package.\n * Functions are exported for use by other stats modules but are not part\n * of the stable public API exported from `src/stats/index.ts`.\n *\n * Some functions (particularly CDFs and special functions like `normalCdf`,\n * `studentTCdf`, `logGamma`, etc.) may be promoted to the public API in\n * future versions if there is user demand.\n *\n * @internal\n * @module stats/_internal\n */\n\nimport type { Axis, Shape } from \"../core\";\nimport {\n  DTypeError,\n  getElementAsNumber,\n  InvalidParameterError,\n  normalizeAxis,\n  ShapeError,\n  validateShape,\n} from \"../core\";\n\nimport { Tensor } from \"../ndarray\";\n\n/**\n * Type representing axis specification for reduction operations.\n * Can be a single axis number/alias or an array of axis numbers/aliases.\n *\n * @example\n * ```ts\n * const axis1: AxisLike = 0;        // Single axis\n * const axis2: AxisLike = [0, 1];   // Multiple axes\n * const axis3: AxisLike = \"rows\";   // Alias\n * ```\n */\nexport type AxisLike = Axis | readonly Axis[];\n\n/**\n * Normalizes axis specification to a sorted array of non-negative axis indices.\n *\n * Converts negative indices to positive, validates bounds, removes duplicates,\n * and returns a sorted array. Returns empty array if axis is undefined.\n *\n * @param axis - Axis specification (single number, array, or undefined)\n * @param ndim - Number of dimensions in the tensor\n * @returns Sorted array of unique, non-negative axis indices\n * @throws {RangeError} If any axis is out of bounds for the given ndim\n *\n * @example\n * ```ts\n * normalizeAxes([-1], 3);      // Returns [2]\n * normalizeAxes([1, 0, 1], 3); // Returns [0, 1] (sorted, deduplicated)\n * normalizeAxes(undefined, 3); // Returns []\n * ```\n */\nexport function normalizeAxes(axis: AxisLike | undefined, ndim: number): readonly number[] {\n  if (axis === undefined) return [];\n  const axesInput: Axis[] = Array.isArray(axis) ? [...axis] : [axis];\n\n  const seen = new Set<number>();\n  const result: number[] = [];\n\n  for (const ax of axesInput) {\n    const norm = normalizeAxis(ax, ndim);\n    if (!seen.has(norm)) {\n      seen.add(norm);\n      result.push(norm);\n    }\n  }\n\n  // Sort for consistency with previous behavior\n  return result.sort((a, b) => a - b);\n}\n\n/**\n * Computes the output shape after reduction along specified axes.\n *\n * When keepdims=true, reduced dimensions become 1.\n * When keepdims=false, reduced dimensions are removed entirely.\n *\n * @param shape - Original tensor shape\n * @param axes - Axes to reduce over (must be normalized)\n * @param keepdims - Whether to keep reduced dimensions as size 1\n * @returns New shape after reduction\n * @throws {ShapeError} If shape dimensions are invalid\n *\n * @example\n * ```ts\n * reducedShape([3, 4, 5], [1], false);    // [3, 5]\n * reducedShape([3, 4, 5], [1], true);     // [3, 1, 5]\n * reducedShape([3, 4, 5], [], false);     // []\n * reducedShape([3, 4, 5], [], true);      // [1, 1, 1]\n * ```\n */\nexport function reducedShape(shape: Shape, axes: readonly number[], keepdims: boolean): Shape {\n  if (axes.length === 0) {\n    return keepdims ? new Array<number>(shape.length).fill(1) : [];\n  }\n\n  const reduce = new Set(axes);\n  const out: number[] = [];\n\n  for (let i = 0; i < shape.length; i++) {\n    const d = shape[i];\n    if (d === undefined) throw new ShapeError(\"Internal error: missing shape dimension\");\n\n    if (reduce.has(i)) {\n      if (keepdims) out.push(1);\n    } else {\n      out.push(d);\n    }\n  }\n\n  validateShape(out);\n  return out;\n}\n\n/**\n * Computes row-major (C-order) strides for a given shape.\n *\n * Strides define how many elements to skip in memory to move one position\n * along each dimension. Computed in row-major order where the last dimension\n * is contiguous in memory.\n *\n * @param shape - Tensor shape\n * @returns Array of strides for each dimension\n *\n * @example\n * ```ts\n * computeStrides([3, 4, 5]); // Returns [20, 5, 1]\n * computeStrides([2, 3]);    // Returns [3, 1]\n * ```\n */\nexport function computeStrides(shape: readonly number[]): readonly number[] {\n  const strides = new Array<number>(shape.length);\n  let stride = 1;\n  for (let i = shape.length - 1; i >= 0; i--) {\n    strides[i] = stride;\n    stride *= shape[i] ?? 0;\n  }\n  return strides;\n}\n\n/**\n * Asserts that two tensors have the same total number of elements.\n *\n * Used for operations that require element-wise correspondence between tensors,\n * regardless of their shapes (e.g., correlation between flattened arrays).\n *\n * @param a - First tensor\n * @param b - Second tensor\n * @param name - Name of the calling function (for error messages)\n * @throws {InvalidParameterError} If tensors have different sizes\n *\n * @example\n * ```ts\n * assertSameSize(tensor([1, 2, 3]), tensor([4, 5, 6]), \"pearsonr\"); // OK\n * assertSameSize(tensor([1, 2]), tensor([3, 4, 5]), \"pearsonr\");    // Throws\n * ```\n */\nexport function assertSameSize(a: Tensor, b: Tensor, name: string): void {\n  if (a.size !== b.size) {\n    throw new InvalidParameterError(\n      `${name}: tensors must have the same number of elements; got ${a.size} and ${b.size}`,\n      \"size\",\n      { a: a.size, b: b.size }\n    );\n  }\n}\n\n/**\n * Extracts a numeric value from a tensor at a specific memory offset.\n *\n * Handles type conversion from bigint to number and validates dtype.\n * This is a low-level accessor used by reduction operations.\n *\n * @param t - Tensor to read from\n * @param offset - Memory offset in the underlying data array\n * @returns Numeric value at the offset\n * @throws {DTypeError} If tensor has string dtype\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3]);\n * getNumberAt(t, t.offset + 1); // Returns 2\n * ```\n */\nexport function getNumberAt(t: Tensor, offset: number): number {\n  if (t.dtype === \"string\" || Array.isArray(t.data)) {\n    throw new DTypeError(\"operation not supported for string dtype\");\n  }\n\n  return getElementAsNumber(t.data, offset);\n}\n\n/**\n * Assigns average ranks to values with tie tracking.\n *\n * Ranks are 1-indexed. Ties receive the average of their rank positions.\n * Returns the sum of (t^3 - t) over tied groups, which is used for tie correction.\n *\n * @param values - Input values to rank\n * @returns Object containing ranks and tie sum\n */\nexport function rankData(values: Float64Array): {\n  ranks: Float64Array;\n  tieSum: number;\n} {\n  const n = values.length;\n  const ranks = new Float64Array(n);\n  if (n === 0) return { ranks, tieSum: 0 };\n\n  const sorted = Array.from(values, (v, i) => ({ v, i })).sort((a, b) => a.v - b.v);\n  let tieSum = 0;\n\n  for (let i = 0; i < sorted.length; ) {\n    let j = i + 1;\n    while (j < sorted.length && sorted[j]?.v === sorted[i]?.v) {\n      j++;\n    }\n    const t = j - i;\n    const avgRank = (i + 1 + j) / 2;\n    for (let k = i; k < j; k++) {\n      const idx = sorted[k]?.i;\n      if (idx !== undefined) {\n        ranks[idx] = avgRank;\n      }\n    }\n    if (t > 1) tieSum += t * t * t - t;\n    i = j;\n  }\n\n  return { ranks, tieSum };\n}\n\n/**\n * Iterates over all elements in a tensor, calling a function with offset and index.\n *\n * Handles arbitrary tensor shapes and strides, iterating in row-major order.\n * This is the core iteration primitive used by all reduction operations.\n *\n * @param t - Tensor to iterate over\n * @param fn - Callback function receiving (offset, index) for each element\n *\n * @example\n * ```ts\n * const t = tensor([[1, 2], [3, 4]]);\n * forEachIndexOffset(t, (offset, idx) => {\n *   console.log(`idx=${idx}, value=${t.data[offset]}`);\n * });\n * // Outputs: idx=[0,0], value=1\n * //          idx=[0,1], value=2\n * //          idx=[1,0], value=3\n * //          idx=[1,1], value=4\n * ```\n */\nexport function forEachIndexOffset(\n  t: Tensor,\n  fn: (offset: number, idx: readonly number[]) => void\n): void {\n  if (t.size === 0) return;\n\n  if (t.ndim === 0) {\n    fn(t.offset, []);\n    return;\n  }\n\n  const shape = t.shape;\n  const strides = t.strides;\n  const idx = new Array<number>(t.ndim).fill(0);\n  let offset = t.offset;\n\n  while (true) {\n    fn(offset, idx);\n\n    // Odometer increment from last axis.\n    let axis = t.ndim - 1;\n    for (;;) {\n      idx[axis] = (idx[axis] ?? 0) + 1;\n      offset += strides[axis] ?? 0;\n\n      const dim = shape[axis] ?? 0;\n      if ((idx[axis] ?? 0) < dim) break;\n\n      // carry\n      offset -= (idx[axis] ?? 0) * (strides[axis] ?? 0);\n      idx[axis] = 0;\n      axis--;\n      if (axis < 0) return;\n    }\n  }\n}\n\n/**\n * Computes the arithmetic mean along specified axes.\n *\n * Uses a simple sum-based approach for numerical stability.\n * This is an internal function used by the public mean() API.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes to reduce over (undefined means all)\n * @param keepdims - Whether to keep reduced dimensions as size 1\n * @returns Tensor containing mean values\n * @throws {InvalidParameterError} If tensor is empty or reduction over empty axis\n *\n * @example\n * ```ts\n * const t = tensor([[1, 2], [3, 4]]);\n * reduceMean(t, undefined, false); // Returns tensor([2.5])\n * reduceMean(t, 0, false);         // Returns tensor([2, 3])\n * ```\n */\nexport function reduceMean(t: Tensor, axis: AxisLike | undefined, keepdims: boolean): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n  if (axes.length === 0) {\n    // Full reduction.\n    if (t.size === 0) {\n      throw new InvalidParameterError(\"mean() requires at least one element\", \"size\", t.size);\n    }\n\n    let sum = 0;\n    forEachIndexOffset(t, (off) => {\n      sum += getNumberAt(t, off);\n    });\n\n    const out = new Float64Array(1);\n    out[0] = sum / t.size;\n\n    const outShape = keepdims ? new Array<number>(t.ndim).fill(1) : [];\n    return Tensor.fromTypedArray({\n      data: out,\n      shape: outShape,\n      dtype: \"float64\",\n      device: t.device,\n    });\n  }\n\n  const outShape = reducedShape(t.shape, axes, keepdims);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sums = new Float64Array(outSize);\n\n  const reduce = new Set<number>(axes);\n  const reduceCount = axes.reduce((acc, ax) => acc * (t.shape[ax] ?? 0), 1);\n  if (reduceCount === 0) {\n    throw new InvalidParameterError(\n      \"mean() reduction over empty axis is undefined\",\n      \"reduceCount\",\n      reduceCount\n    );\n  }\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    if (keepdims) {\n      // outShape has same rank as input.\n      for (let i = 0; i < t.ndim; i++) {\n        const s = outStrides[i] ?? 0;\n        const v = reduce.has(i) ? 0 : (idx[i] ?? 0);\n        outFlat += v * s;\n      }\n    } else {\n      let oi = 0;\n      for (let i = 0; i < t.ndim; i++) {\n        if (reduce.has(i)) continue;\n        outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n        oi++;\n      }\n    }\n\n    sums[outFlat] = (sums[outFlat] ?? 0) + getNumberAt(t, off);\n  });\n\n  for (let i = 0; i < sums.length; i++) {\n    sums[i] = (sums[i] ?? 0) / reduceCount;\n  }\n\n  return Tensor.fromTypedArray({\n    data: sums,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes variance along specified axes using Welford's online algorithm.\n *\n * Welford's algorithm provides numerical stability for variance computation\n * by avoiding catastrophic cancellation that occurs with the naive two-pass method.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes to reduce over (undefined means all)\n * @param keepdims - Whether to keep reduced dimensions as size 1\n * @param ddof - Delta degrees of freedom (0 for population, 1 for sample variance)\n * @returns Tensor containing variance values\n * @throws {InvalidParameterError} If tensor is empty, ddof >= sample size, or reduction over empty axis\n * @throws {DTypeError} If tensor has string dtype\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * reduceVariance(t, undefined, false, 1); // Sample variance\n * reduceVariance(t, undefined, false, 0); // Population variance\n * ```\n */\nexport function reduceVariance(\n  t: Tensor,\n  axis: AxisLike | undefined,\n  keepdims: boolean,\n  ddof: number\n): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n\n  if (t.dtype === \"string\") {\n    throw new DTypeError(\"variance() not supported for string dtype\");\n  }\n\n  if (axes.length === 0) {\n    if (t.size === 0) {\n      throw new InvalidParameterError(\"variance() requires at least one element\", \"size\", t.size);\n    }\n    if (ddof < 0) {\n      throw new InvalidParameterError(\"ddof must be non-negative\", \"ddof\", ddof);\n    }\n    if (t.size <= ddof) {\n      throw new InvalidParameterError(\n        `ddof=${ddof} >= size=${t.size}, variance undefined`,\n        \"ddof\",\n        ddof\n      );\n    }\n\n    // Welford's online algorithm for numerically stable variance computation\n    // Maintains running mean and sum of squared deviations (M2)\n    let mean = 0; // Running mean\n    let m2 = 0; // Sum of squared deviations from mean\n    let n = 0; // Count of elements processed\n    forEachIndexOffset(t, (off) => {\n      const x = getNumberAt(t, off);\n      n++;\n      const delta = x - mean; // Deviation from old mean\n      mean += delta / n; // Update mean incrementally\n      const delta2 = x - mean; // Deviation from new mean\n      m2 += delta * delta2; // Update M2 (numerically stable)\n    });\n\n    const out = new Float64Array(1);\n    out[0] = m2 / (n - ddof);\n\n    const outShape = keepdims ? new Array<number>(t.ndim).fill(1) : [];\n    return Tensor.fromTypedArray({\n      data: out,\n      shape: outShape,\n      dtype: \"float64\",\n      device: t.device,\n    });\n  }\n\n  const outShape = reducedShape(t.shape, axes, keepdims);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n\n  const reduce = new Set<number>(axes);\n  const reduceCount = axes.reduce((acc, ax) => acc * (t.shape[ax] ?? 0), 1);\n  if (reduceCount === 0) {\n    throw new InvalidParameterError(\n      \"variance() reduction over empty axis is undefined\",\n      \"reduceCount\",\n      reduceCount\n    );\n  }\n  if (ddof < 0) {\n    throw new InvalidParameterError(\"ddof must be non-negative\", \"ddof\", ddof);\n  }\n  if (reduceCount <= ddof) {\n    throw new InvalidParameterError(\n      `ddof=${ddof} >= reduced size=${reduceCount}, variance undefined`,\n      \"ddof\",\n      ddof\n    );\n  }\n\n  const means = new Float64Array(outSize);\n  const m2s = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    if (keepdims) {\n      for (let i = 0; i < t.ndim; i++) {\n        const s = outStrides[i] ?? 0;\n        const v = reduce.has(i) ? 0 : (idx[i] ?? 0);\n        outFlat += v * s;\n      }\n    } else {\n      let oi = 0;\n      for (let i = 0; i < t.ndim; i++) {\n        if (reduce.has(i)) continue;\n        outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n        oi++;\n      }\n    }\n\n    const x = getNumberAt(t, off);\n    const n = (counts[outFlat] ?? 0) + 1;\n    counts[outFlat] = n;\n\n    const mean = means[outFlat] ?? 0;\n    const delta = x - mean;\n    const nextMean = mean + delta / n;\n    means[outFlat] = nextMean;\n    const delta2 = x - nextMean;\n    m2s[outFlat] = (m2s[outFlat] ?? 0) + delta * delta2;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const n = counts[i] ?? 0;\n    out[i] = (m2s[i] ?? 0) / (n - ddof);\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n// ---- Special functions for distribution CDFs ----\n\n/**\n * Lanczos approximation coefficients for gamma function (g=7, n=9).\n * These constants provide high-precision approximation of the gamma function.\n */\nconst LANCZOS_COEFFS: readonly number[] = [\n  0.99999999999980993, 676.5203681218851, -1259.1392167224028, 771.3234287776531,\n  -176.6150291621406, 12.507343278686905, -0.13857109526572012, 0.000009984369578019572,\n  1.5056327351493116e-7,\n];\n\n/**\n * Computes the natural logarithm of the gamma function: ln(Γ(z)).\n *\n * Uses Lanczos approximation for z >= 0.5 and reflection formula for z < 0.5.\n * The gamma function extends the factorial to real and complex numbers:\n * Γ(n) = (n-1)! for positive integers n.\n *\n * @param z - Input value (real number)\n * @returns Natural logarithm of gamma function at z\n *\n * @example\n * ```ts\n * logGamma(5);    // Returns ln(4!) = ln(24) ≈ 3.178\n * logGamma(0.5);  // Returns ln(√π) ≈ 0.572\n * ```\n */\nexport function logGamma(z: number): number {\n  if (z < 0.5) {\n    // Use reflection formula: Γ(z)Γ(1-z) = π/sin(πz)\n    // Therefore: ln(Γ(z)) = ln(π) - ln(sin(πz)) - ln(Γ(1-z))\n    return Math.log(Math.PI) - Math.log(Math.sin(Math.PI * z)) - logGamma(1 - z);\n  }\n\n  // Lanczos approximation for z >= 0.5\n  z -= 1; // Shift z for Lanczos formula\n  let x = LANCZOS_COEFFS[0] ?? 0; // Start with first coefficient\n  // Sum the series: x = c0 + c1/(z+1) + c2/(z+2) + ... + c8/(z+8)\n  for (let i = 1; i < LANCZOS_COEFFS.length; i++) {\n    x += (LANCZOS_COEFFS[i] ?? 0) / (z + i);\n  }\n\n  const t = z + 7.5; // g + 0.5 where g=7\n  // Final Lanczos formula: ln(Γ(z+1)) = 0.5*ln(2π) + (z+0.5)*ln(t) - t + ln(x)\n  return 0.5 * Math.log(2 * Math.PI) + (z + 0.5) * Math.log(t) - t + Math.log(x);\n}\n\n/**\n * Evaluates continued fraction for incomplete beta function.\n *\n * Uses Lentz's algorithm for evaluating continued fractions.\n * This is a helper function for regularizedIncompleteBeta.\n *\n * @param a - First shape parameter\n * @param b - Second shape parameter\n * @param x - Evaluation point in [0, 1]\n * @returns Continued fraction value\n */\nfunction betacf(a: number, b: number, x: number): number {\n  const MAX_ITER = 200; // Maximum iterations for convergence\n  const EPS = 3e-14; // Convergence threshold\n  const FPMIN = 1e-300; // Minimum floating point value to prevent division by zero\n\n  // Precompute common terms\n  const qab = a + b;\n  const qap = a + 1;\n  const qam = a - 1;\n\n  // Initialize Lentz's algorithm\n  let c = 1;\n  let d = 1 - (qab * x) / qap;\n  if (Math.abs(d) < FPMIN) d = FPMIN; // Prevent division by zero\n  d = 1 / d;\n  let h = d; // Accumulated result\n\n  // Iterate using modified Lentz's method\n  for (let m = 1; m <= MAX_ITER; m++) {\n    const m2 = 2 * m;\n\n    // Even step of continued fraction\n    let aa = (m * (b - m) * x) / ((qam + m2) * (a + m2));\n    d = 1 + aa * d;\n    if (Math.abs(d) < FPMIN) d = FPMIN;\n    c = 1 + aa / c;\n    if (Math.abs(c) < FPMIN) c = FPMIN;\n    d = 1 / d;\n    h *= d * c;\n\n    // Odd step of continued fraction\n    aa = (-(a + m) * (qab + m) * x) / ((a + m2) * (qap + m2));\n    d = 1 + aa * d;\n    if (Math.abs(d) < FPMIN) d = FPMIN;\n    c = 1 + aa / c;\n    if (Math.abs(c) < FPMIN) c = FPMIN;\n    d = 1 / d;\n    const del = d * c;\n    h *= del;\n\n    // Check for convergence\n    if (Math.abs(del - 1.0) < EPS) break;\n  }\n\n  return h;\n}\n\n/**\n * Computes the regularized incomplete beta function I_x(a, b).\n *\n * The regularized incomplete beta function is defined as:\n * I_x(a, b) = B(x; a, b) / B(a, b)\n * where B(x; a, b) is the incomplete beta function and B(a, b) is the beta function.\n *\n * Used in computing CDFs for beta, F, and t distributions.\n *\n * @param a - First shape parameter (must be > 0)\n * @param b - Second shape parameter (must be > 0)\n * @param x - Evaluation point in [0, 1]\n * @returns Value of I_x(a, b) in [0, 1]\n * @throws {InvalidParameterError} If parameters are outside their valid ranges\n *\n * @example\n * ```ts\n * regularizedIncompleteBeta(2, 3, 0.5); // Returns ~0.6875\n * ```\n */\nexport function regularizedIncompleteBeta(a: number, b: number, x: number): number {\n  if (!Number.isFinite(a) || a <= 0) {\n    throw new InvalidParameterError(\"a must be > 0\", \"a\", a);\n  }\n  if (!Number.isFinite(b) || b <= 0) {\n    throw new InvalidParameterError(\"b must be > 0\", \"b\", b);\n  }\n  // Validate input range\n  if (x < 0 || x > 1) {\n    throw new InvalidParameterError(\"x must be in [0,1]\", \"x\", x);\n  }\n  // Handle boundary cases\n  if (x === 0) return 0;\n  if (x === 1) return 1;\n\n  // Compute ln(B(x; a, b)) = ln(Γ(a+b)) - ln(Γ(a)) - ln(Γ(b)) + a*ln(x) + b*ln(1-x)\n  const lnBt = logGamma(a + b) - logGamma(a) - logGamma(b) + a * Math.log(x) + b * Math.log(1 - x);\n  const bt = Math.exp(lnBt);\n\n  // Use symmetry relation to ensure x is in the more stable region\n  if (x < (a + 1) / (a + b + 2)) {\n    // Direct evaluation\n    return (bt * betacf(a, b, x)) / a;\n  }\n  // Use symmetry: I_x(a,b) = 1 - I_(1-x)(b,a)\n  return 1 - (bt * betacf(b, a, 1 - x)) / b;\n}\n\n/**\n * Computes the regularized lower incomplete gamma function P(s, x).\n *\n * P(s, x) = γ(s, x) / Γ(s) where γ(s, x) is the lower incomplete gamma function.\n * This represents the CDF of the gamma distribution.\n *\n * @param s - Shape parameter (must be > 0)\n * @param x - Evaluation point (must be >= 0)\n * @returns Value of P(s, x) in [0, 1]\n * @throws {InvalidParameterError} If parameters are outside their valid ranges\n */\nfunction regularizedLowerIncompleteGamma(s: number, x: number): number {\n  // Validate input\n  if (!Number.isFinite(s) || s <= 0) {\n    throw new InvalidParameterError(\"s must be > 0\", \"s\", s);\n  }\n  if (x < 0) throw new InvalidParameterError(\"x must be >= 0\", \"x\", x);\n  if (x === 0) return 0;\n\n  const ITMAX = 200; // Maximum iterations\n  const EPS = 3e-14; // Convergence threshold\n  const FPMIN = 1e-300; // Minimum floating point value\n\n  if (x < s + 1) {\n    // Use series representation for x < s + 1 (more stable)\n    // P(s,x) = e^(-x) * x^s / Γ(s) * Σ(x^n / Γ(s+n+1))\n    let sum = 1 / s;\n    let del = sum;\n    let ap = s;\n\n    for (let n = 1; n <= ITMAX; n++) {\n      ap += 1;\n      del *= x / ap;\n      sum += del;\n      if (Math.abs(del) < Math.abs(sum) * EPS) break; // Converged\n    }\n\n    // Multiply by normalization factor\n    return sum * Math.exp(-x + s * Math.log(x) - logGamma(s));\n  }\n\n  // Use continued fraction for Q(s, x) = 1 - P(s, x) when x >= s + 1\n  // This is more numerically stable in this region\n  let b = x + 1 - s;\n  let c = 1 / FPMIN;\n  let d = 1 / b;\n  let h = d;\n\n  // Evaluate continued fraction using Lentz's algorithm\n  for (let i = 1; i <= ITMAX; i++) {\n    const an = -i * (i - s);\n    b += 2;\n    d = an * d + b;\n    if (Math.abs(d) < FPMIN) d = FPMIN;\n    c = b + an / c;\n    if (Math.abs(c) < FPMIN) c = FPMIN;\n    d = 1 / d;\n    const del = d * c;\n    h *= del;\n    if (Math.abs(del - 1) < EPS) break; // Converged\n  }\n\n  // Compute Q(s, x) and return P(s, x) = 1 - Q(s, x)\n  const q = Math.exp(-x + s * Math.log(x) - logGamma(s)) * h;\n  return 1 - q;\n}\n\n/**\n * Computes the cumulative distribution function (CDF) of the standard normal distribution.\n *\n * Uses Abramowitz & Stegun approximation via error function.\n * Φ(x) = 0.5 * (1 + erf(x/√2))\n *\n * @param x - Input value\n * @returns Probability P(X <= x) where X ~ N(0, 1)\n *\n * @example\n * ```ts\n * normalCdf(0);     // Returns 0.5\n * normalCdf(1.96);  // Returns ~0.975 (95th percentile)\n * normalCdf(-1.96); // Returns ~0.025 (5th percentile)\n * ```\n */\nexport function normalCdf(x: number): number {\n  // Abramowitz & Stegun approximation via error function\n  // erf(x) ≈ sign(x) * sqrt(1 - exp(-x^2 * (4/π + ax^2) / (1 + ax^2)))\n  const a = 0.147; // Approximation parameter\n  const sign = x < 0 ? -1 : 1;\n  const ax = Math.abs(x) / Math.SQRT2; // Scale by 1/√2 for erf\n  const t = 1 + a * ax * ax;\n  const erf = sign * Math.sqrt(1 - Math.exp(-ax * ax * ((4 / Math.PI + a * ax * ax) / t)));\n  // Convert erf to CDF: Φ(x) = 0.5 * (1 + erf(x/√2))\n  return 0.5 * (1 + erf);\n}\n\n/**\n * Computes the cumulative distribution function (CDF) of Student's t-distribution.\n *\n * Uses the relationship between t-distribution and incomplete beta function:\n * F_t(t; ν) = 0.5 * I_x(ν/2, 1/2) where x = ν/(ν + t²)\n *\n * @param t - t-statistic value\n * @param df - Degrees of freedom (must be > 0)\n * @returns Probability P(T <= t) where T ~ t(df)\n * @throws {InvalidParameterError} If df <= 0\n *\n * @example\n * ```ts\n * studentTCdf(0, 10);      // Returns 0.5 (symmetric at 0)\n * studentTCdf(2.228, 10);  // Returns ~0.975 (95th percentile for df=10)\n * ```\n */\nexport function studentTCdf(t: number, df: number): number {\n  // Validate degrees of freedom\n  if (!Number.isFinite(df) || df <= 0) {\n    throw new InvalidParameterError(\"df must be > 0\", \"df\", df);\n  }\n  // Handle infinite t-values\n  if (Number.isNaN(t)) return NaN;\n  if (!Number.isFinite(t)) return t < 0 ? 0 : 1;\n\n  // Transform to incomplete beta function parameter\n  const x = df / (df + t * t);\n  const a = df / 2;\n  const b = 0.5;\n  const ib = regularizedIncompleteBeta(a, b, x);\n  const p = 0.5 * ib;\n  // Use symmetry of t-distribution around 0\n  return t >= 0 ? 1 - p : p;\n}\n\n/**\n * Computes the cumulative distribution function (CDF) of the chi-square distribution.\n *\n * Uses the relationship: χ²(x; k) = P(k/2, x/2)\n * where P is the regularized lower incomplete gamma function.\n *\n * @param x - Chi-square statistic (must be >= 0)\n * @param k - Degrees of freedom (must be > 0)\n * @returns Probability P(X <= x) where X ~ χ²(k)\n * @throws {InvalidParameterError} If k <= 0\n *\n * @example\n * ```ts\n * chiSquareCdf(3.841, 1);  // Returns ~0.95 (95th percentile for df=1)\n * chiSquareCdf(0, 5);      // Returns 0\n * ```\n */\nexport function chiSquareCdf(x: number, k: number): number {\n  // Validate degrees of freedom\n  if (!Number.isFinite(k) || k <= 0) {\n    throw new InvalidParameterError(\"degrees of freedom must be > 0\", \"k\", k);\n  }\n  if (Number.isNaN(x)) return NaN;\n  if (x === Infinity) return 1;\n  // Chi-square is non-negative\n  if (x <= 0) return 0;\n  // Use gamma CDF relationship: χ²(k) is Gamma(k/2, 2)\n  return regularizedLowerIncompleteGamma(k / 2, x / 2);\n}\n\n/**\n * Computes the cumulative distribution function (CDF) of the F-distribution.\n *\n * Uses the relationship between F-distribution and incomplete beta function:\n * F(x; d₁, d₂) = I_y(d₁/2, d₂/2) where y = (d₁*x)/(d₁*x + d₂)\n *\n * @param x - F-statistic value (must be >= 0)\n * @param dfn - Numerator degrees of freedom (must be > 0)\n * @param dfd - Denominator degrees of freedom (must be > 0)\n * @returns Probability P(F <= x) where F ~ F(dfn, dfd)\n * @throws {InvalidParameterError} If dfn <= 0 or dfd <= 0\n *\n * @example\n * ```ts\n * fCdf(4.0, 5, 10);  // Returns F-distribution CDF at x=4\n * fCdf(0, 5, 10);    // Returns 0\n * ```\n */\nexport function fCdf(x: number, dfn: number, dfd: number): number {\n  // Validate degrees of freedom\n  if (!Number.isFinite(dfn) || dfn <= 0) {\n    throw new InvalidParameterError(\"degrees of freedom (dfn) must be > 0\", \"dfn\", dfn);\n  }\n  if (!Number.isFinite(dfd) || dfd <= 0) {\n    throw new InvalidParameterError(\"degrees of freedom (dfd) must be > 0\", \"dfd\", dfd);\n  }\n  if (Number.isNaN(x)) return NaN;\n  if (x === Infinity) return 1;\n  // F-statistic is non-negative\n  if (x <= 0) return 0;\n\n  // Transform to incomplete beta parameter\n  const xx = (dfn * x) / (dfn * x + dfd);\n  return regularizedIncompleteBeta(dfn / 2, dfd / 2, xx);\n}\n","import { InvalidParameterError, ShapeError } from \"../core\";\nimport { Tensor, tensor } from \"../ndarray\";\nimport {\n  assertSameSize,\n  forEachIndexOffset,\n  getNumberAt,\n  normalCdf,\n  rankData,\n  studentTCdf,\n} from \"./_internal\";\n\n/**\n * Converts a tensor to a dense flat Float64Array.\n * Helper function for correlation computations.\n *\n * @param t - Input tensor\n * @returns Flattened Float64Array containing all tensor values\n */\nfunction toDenseFlatArray(t: Tensor): Float64Array {\n  const out = new Float64Array(t.size);\n  let i = 0;\n  forEachIndexOffset(t, (off) => {\n    out[i] = getNumberAt(t, off);\n    i++;\n  });\n  return out;\n}\n\n/**\n * Computes Pearson correlation coefficient from two dense arrays.\n *\n * Uses the standard formula: r = cov(X,Y) / (std(X) * std(Y))\n * Computed efficiently in a single pass.\n *\n * @param x - First array\n * @param y - Second array (must have same length as x)\n * @returns Pearson correlation coefficient in [-1, 1]\n * @throws {InvalidParameterError} If arrays have constant values (zero variance)\n */\nfunction pearsonFromDense(x: Float64Array, y: Float64Array): number {\n  const n = x.length;\n  // First pass: compute means\n  let sumX = 0;\n  let sumY = 0;\n  for (let i = 0; i < n; i++) {\n    sumX += x[i] ?? 0;\n    sumY += y[i] ?? 0;\n  }\n  const meanX = sumX / n;\n  const meanY = sumY / n;\n\n  // Second pass: compute covariance and variances\n  let num = 0; // Numerator: covariance\n  let denX = 0; // Denominator: variance of X\n  let denY = 0; // Denominator: variance of Y\n  for (let i = 0; i < n; i++) {\n    const dx = (x[i] ?? 0) - meanX;\n    const dy = (y[i] ?? 0) - meanY;\n    num += dx * dy; // Sum of products of deviations\n    denX += dx * dx; // Sum of squared deviations for X\n    denY += dy * dy; // Sum of squared deviations for Y\n  }\n\n  // Compute correlation: r = cov(X,Y) / (std(X) * std(Y))\n  const den = Math.sqrt(denX * denY);\n  if (den === 0) {\n    throw new InvalidParameterError(\n      \"pearsonr() is undefined for constant input\",\n      \"input\",\n      \"constant\"\n    );\n  }\n  return num / den;\n}\n\n/**\n * Pearson correlation coefficient.\n *\n * Measures linear correlation between two variables.\n *\n * @param x - First tensor\n * @param y - Second tensor (must have same size as x)\n * @returns Tuple of [correlation coefficient in [-1, 1], two-tailed p-value]\n * @throws {InvalidParameterError} If tensors have different sizes, < 2 samples, or constant input\n *\n * @example\n * ```ts\n * const x = tensor([1, 2, 3, 4, 5]);\n * const y = tensor([2, 4, 6, 8, 10]);\n * const [r, p] = pearsonr(x, y);  // r = 1.0 (perfect linear)\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN correlation\n * - Infinity inputs result in NaN correlation\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.pearsonr.html | SciPy stats.pearsonr}\n */\nexport function pearsonr(x: Tensor, y: Tensor): [number, number] {\n  assertSameSize(x, y, \"pearsonr\");\n  const n = x.size;\n  if (n < 2) {\n    throw new InvalidParameterError(\"pearsonr() requires at least 2 paired samples\", \"n\", n);\n  }\n\n  const xd = toDenseFlatArray(x);\n  const yd = toDenseFlatArray(y);\n  const r = pearsonFromDense(xd, yd);\n\n  // Compute p-value using t-distribution\n  // Under H0: ρ=0, the test statistic t = r*sqrt((n-2)/(1-r²)) follows t(n-2)\n  const df = n - 2;\n  if (df <= 0) {\n    return [r, NaN]; // Cannot compute p-value with < 2 degrees of freedom\n  }\n  const tStat = r * Math.sqrt(df / (1 - r * r));\n  const pValue = 2 * (1 - studentTCdf(Math.abs(tStat), df)); // Two-tailed test\n  return [r, pValue];\n}\n\n/**\n * Computes Spearman's rank correlation coefficient.\n *\n * Non-parametric measure of monotonic relationship between two variables.\n * Computed as Pearson correlation of rank values.\n * - ρ = 1: Perfect monotonic increasing relationship\n * - ρ = 0: No monotonic relationship\n * - ρ = -1: Perfect monotonic decreasing relationship\n *\n * @param x - First tensor\n * @param y - Second tensor (must have same size as x)\n * @returns Tuple of [correlation coefficient, p-value]\n * @throws {InvalidParameterError} If tensors have different sizes, < 2 samples, or constant input\n *\n * @example\n * ```ts\n * const x = tensor([1, 2, 3, 4, 5]);\n * const y = tensor([2, 4, 6, 8, 10]);\n * const [rho, p] = spearmanr(x, y);  // rho = 1.0 (perfect monotonic)\n * ```\n *\n * @remarks\n * Ties are assigned average ranks.\n * NaN values are ranked according to JavaScript sort behavior.\n * Infinity values are sorted naturally (±Infinity at extremes).\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.spearmanr.html | SciPy stats.spearmanr}\n */\nexport function spearmanr(x: Tensor, y: Tensor): [number, number] {\n  assertSameSize(x, y, \"spearmanr\");\n  const n = x.size;\n  if (n < 2) {\n    throw new InvalidParameterError(\"spearmanr() requires at least 2 paired samples\", \"n\", n);\n  }\n\n  const xd = toDenseFlatArray(x);\n  const yd = toDenseFlatArray(y);\n\n  // Convert values to average ranks (1-indexed) with tie correction.\n  const rankX = rankData(xd).ranks;\n  const rankY = rankData(yd).ranks;\n\n  // Compute Pearson correlation of ranks\n  const rho = pearsonFromDense(rankX, rankY);\n  // Test statistic follows t-distribution under H0: ρ=0\n  const df = n - 2;\n  const tStat = rho * Math.sqrt(df / (1 - rho * rho));\n  const pValue = df > 0 ? 2 * (1 - studentTCdf(Math.abs(tStat), df)) : NaN;\n  return [rho, pValue];\n}\n\n/**\n * Computes Kendall's tau correlation coefficient.\n *\n * Non-parametric measure of ordinal association based on concordant/discordant pairs.\n * More robust to outliers than Spearman, but computationally more expensive.\n * - τ = 1: All pairs concordant (perfect agreement)\n * - τ = 0: Equal concordant and discordant pairs\n * - τ = -1: All pairs discordant (perfect disagreement)\n *\n * @param x - First tensor\n * @param y - Second tensor (must have same size as x)\n * @returns Tuple of [tau coefficient, p-value]\n * @throws {InvalidParameterError} If tensors have different sizes or < 2 samples\n *\n * @example\n * ```ts\n * const x = tensor([1, 2, 3, 4, 5]);\n * const y = tensor([1, 3, 2, 4, 5]);\n * const [tau, p] = kendalltau(x, y);  // Mostly concordant\n * ```\n *\n * @remarks\n * This implementation uses the tau-b variant with tie correction.\n * Ties are excluded from concordant/discordant counts and reduce the denominator.\n * The p-value uses a normal approximation with tie-corrected variance.\n *\n * @complexity O(n²) - suitable for moderate sample sizes (n < 10,000)\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kendalltau.html | SciPy stats.kendalltau}\n */\nexport function kendalltau(x: Tensor, y: Tensor): [number, number] {\n  assertSameSize(x, y, \"kendalltau\");\n  const n = x.size;\n  if (n < 2) {\n    throw new InvalidParameterError(\"kendalltau() requires at least 2 paired samples\", \"n\", n);\n  }\n\n  const xd = toDenseFlatArray(x);\n  const yd = toDenseFlatArray(y);\n\n  // Count concordant and discordant pairs (ties excluded)\n  let concordant = 0;\n  let discordant = 0;\n  for (let i = 0; i < n - 1; i++) {\n    const xi = xd[i] ?? 0;\n    const yi = yd[i] ?? 0;\n    for (let j = i + 1; j < n; j++) {\n      const signX = Math.sign((xd[j] ?? 0) - xi);\n      const signY = Math.sign((yd[j] ?? 0) - yi);\n      if (signX === 0 || signY === 0) continue;\n      if (signX === signY) concordant++;\n      else discordant++;\n    }\n  }\n\n  const n0 = (n * (n - 1)) / 2;\n  // Tie summaries for tau-b denominator and variance corrections.\n  const tieSums = (\n    vals: Float64Array\n  ): {\n    nTies: number;\n    sumT: number;\n    sumT2: number;\n    sumT3: number;\n  } => {\n    const sorted = Array.from(vals).sort((a, b) => a - b);\n    let sumT = 0;\n    let sumT2 = 0;\n    let sumT3 = 0;\n    for (let i = 0; i < sorted.length; ) {\n      let j = i + 1;\n      while (j < sorted.length && sorted[j] === sorted[i]) j++;\n      const t = j - i;\n      if (t > 1) {\n        sumT += t * (t - 1);\n        sumT2 += t * (t - 1) * (2 * t + 5);\n        sumT3 += t * (t - 1) * (t - 2);\n      }\n      i = j;\n    }\n    return { nTies: sumT / 2, sumT, sumT2, sumT3 };\n  };\n\n  const tieX = tieSums(xd);\n  const tieY = tieSums(yd);\n  const denom = Math.sqrt((n0 - tieX.nTies) * (n0 - tieY.nTies));\n  const s = concordant - discordant;\n  const tau = denom === 0 ? NaN : s / denom;\n\n  // Normal approximation for p-value with tie correction\n  // Normal approximation variance with tie correction (SciPy-style).\n  let varS =\n    (n * (n - 1) * (2 * n + 5) - tieX.sumT2 - tieY.sumT2) / 18 +\n    (tieX.sumT * tieY.sumT) / (2 * n * (n - 1));\n  if (n > 2) {\n    varS += (tieX.sumT3 * tieY.sumT3) / (9 * n * (n - 1) * (n - 2));\n  }\n\n  const pValue = varS <= 0 ? NaN : 2 * (1 - normalCdf(Math.abs(s / Math.sqrt(varS))));\n  return [tau, pValue];\n}\n\n/**\n * Computes the Pearson correlation coefficient matrix.\n *\n * For two variables, returns 2x2 correlation matrix.\n * For a 2D tensor, treats each column as a variable and computes pairwise correlations.\n *\n * @param x - Input tensor (1D or 2D)\n * @param y - Optional second tensor (if provided, computes correlation between x and y)\n * @returns Correlation matrix (symmetric with 1s on diagonal)\n * @throws {InvalidParameterError} If < 2 observations or size mismatch\n * @throws {ShapeError} If tensor is not 1D or 2D\n *\n * @example\n * ```ts\n * const x = tensor([1, 2, 3, 4, 5]);\n * const y = tensor([2, 4, 5, 4, 5]);\n * corrcoef(x, y);  // Returns [[1.0, 0.8], [0.8, 1.0]]\n *\n * const data = tensor([[1, 2], [3, 4], [5, 6]]);\n * corrcoef(data);  // Returns 2x2 correlation matrix for 2 variables\n * ```\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.corrcoef.html | NumPy corrcoef}\n */\nexport function corrcoef(x: Tensor, y?: Tensor): Tensor {\n  if (y) {\n    const [r] = pearsonr(x, y);\n    return tensor([\n      [1.0, r],\n      [r, 1.0],\n    ]);\n  }\n\n  if (x.ndim === 1) {\n    if (x.size < 2) {\n      throw new InvalidParameterError(\n        \"corrcoef() requires at least 2 observations\",\n        \"nObs\",\n        x.size\n      );\n    }\n    return tensor([[1.0]]);\n  }\n\n  if (x.ndim !== 2) {\n    throw new ShapeError(\"corrcoef() expects a 1D or 2D tensor\");\n  }\n\n  // Treat columns as variables (rowvar=false style): shape (nObs, nVar)\n  const nObs = x.shape[0] ?? 0; // Number of observations (rows)\n  const nVar = x.shape[1] ?? 0; // Number of variables (columns)\n  if (nObs < 2) {\n    throw new InvalidParameterError(\"corrcoef() requires at least 2 observations\", \"nObs\", nObs);\n  }\n\n  const s0 = x.strides[0] ?? 0;\n  const s1 = x.strides[1] ?? 0;\n  const xOff = x.offset;\n  const xData = x.data;\n\n  // Fast path: direct numeric array access (non-BigInt, non-string)\n  const directAccess = !Array.isArray(xData) && !(xData instanceof BigInt64Array);\n\n  const means = new Float64Array(nVar);\n  if (directAccess) {\n    for (let j = 0; j < nVar; j++) {\n      let s = 0;\n      for (let i = 0; i < nObs; i++) {\n        s += xData[xOff + i * s0 + j * s1] as number;\n      }\n      means[j] = s / nObs;\n    }\n  } else {\n    for (let j = 0; j < nVar; j++) {\n      let s = 0;\n      for (let i = 0; i < nObs; i++) {\n        s += getNumberAt(x, xOff + i * s0 + j * s1);\n      }\n      means[j] = s / nObs;\n    }\n  }\n\n  const cov = new Float64Array(nVar * nVar);\n  const ddof = 1;\n  if (directAccess) {\n    for (let a = 0; a < nVar; a++) {\n      const ma = means[a] as number;\n      for (let b = a; b < nVar; b++) {\n        const mb = means[b] as number;\n        let s = 0;\n        for (let i = 0; i < nObs; i++) {\n          const base = xOff + i * s0;\n          s += ((xData[base + a * s1] as number) - ma) * ((xData[base + b * s1] as number) - mb);\n        }\n        const v = s / (nObs - ddof);\n        cov[a * nVar + b] = v;\n        cov[b * nVar + a] = v;\n      }\n    }\n  } else {\n    for (let a = 0; a < nVar; a++) {\n      for (let b = a; b < nVar; b++) {\n        let s = 0;\n        for (let i = 0; i < nObs; i++) {\n          const offA = xOff + i * s0 + a * s1;\n          const offB = xOff + i * s0 + b * s1;\n          s += (getNumberAt(x, offA) - (means[a] ?? 0)) * (getNumberAt(x, offB) - (means[b] ?? 0));\n        }\n        const v = s / (nObs - ddof);\n        cov[a * nVar + b] = v;\n        cov[b * nVar + a] = v;\n      }\n    }\n  }\n\n  const corr = new Float64Array(nVar * nVar);\n  for (let i = 0; i < nVar; i++) {\n    for (let j = 0; j < nVar; j++) {\n      const v = cov[i * nVar + j] ?? 0;\n      const vi = cov[i * nVar + i] ?? 0;\n      const vj = cov[j * nVar + j] ?? 0;\n      const den = Math.sqrt(vi * vj);\n      corr[i * nVar + j] = den === 0 ? NaN : v / den;\n    }\n  }\n\n  return Tensor.fromTypedArray({\n    data: corr,\n    shape: [nVar, nVar],\n    dtype: \"float64\",\n    device: x.device,\n  });\n}\n\n/**\n * Computes the covariance matrix.\n *\n * Covariance measures how two variables change together.\n * For two variables, returns 2x2 covariance matrix.\n * For a 2D tensor, treats each column as a variable.\n *\n * @param x - Input tensor (1D or 2D)\n * @param y - Optional second tensor (if provided, computes covariance between x and y)\n * @param ddof - Delta degrees of freedom (0 = population, 1 = sample, default: 1)\n * @returns Covariance matrix (symmetric)\n * @throws {InvalidParameterError} If tensor is empty, ddof < 0, ddof >= sample size, or size mismatch\n * @throws {ShapeError} If tensor is not 1D or 2D\n *\n * @example\n * ```ts\n * const x = tensor([1, 2, 3, 4, 5]);\n * const y = tensor([2, 4, 5, 4, 5]);\n * cov(x, y);  // Returns 2x2 covariance matrix\n *\n * const data = tensor([[1, 2], [3, 4], [5, 6]]);\n * cov(data);  // Returns 2x2 covariance matrix for 2 variables\n * ```\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.cov.html | NumPy cov}\n */\nexport function cov(x: Tensor, y?: Tensor, ddof = 1): Tensor {\n  if (!Number.isFinite(ddof) || ddof < 0) {\n    throw new InvalidParameterError(\"ddof must be a non-negative finite number\", \"ddof\", ddof);\n  }\n\n  if (y) {\n    assertSameSize(x, y, \"cov\");\n    const n = x.size;\n    if (n === 0) throw new InvalidParameterError(\"cov() requires at least one element\", \"n\", n);\n    if (n <= ddof)\n      throw new InvalidParameterError(\n        `ddof=${ddof} >= size=${n}, covariance undefined`,\n        \"ddof\",\n        ddof\n      );\n\n    const xd = toDenseFlatArray(x);\n    const yd = toDenseFlatArray(y);\n\n    let meanX = 0;\n    let meanY = 0;\n    for (let i = 0; i < n; i++) {\n      meanX += xd[i] ?? 0;\n      meanY += yd[i] ?? 0;\n    }\n    meanX /= n;\n    meanY /= n;\n\n    let varX = 0;\n    let varY = 0;\n    let covXY = 0;\n    for (let i = 0; i < n; i++) {\n      const dx = (xd[i] ?? 0) - meanX;\n      const dy = (yd[i] ?? 0) - meanY;\n      varX += dx * dx;\n      varY += dy * dy;\n      covXY += dx * dy;\n    }\n    varX /= n - ddof;\n    varY /= n - ddof;\n    covXY /= n - ddof;\n\n    return tensor([\n      [varX, covXY],\n      [covXY, varY],\n    ]);\n  }\n\n  if (x.ndim === 1) {\n    const n = x.size;\n    if (n === 0) throw new InvalidParameterError(\"cov() requires at least one element\", \"n\", n);\n    if (ddof < 0) {\n      throw new InvalidParameterError(\"ddof must be non-negative\", \"ddof\", ddof);\n    }\n    if (n <= ddof)\n      throw new InvalidParameterError(\n        `ddof=${ddof} >= size=${n}, covariance undefined`,\n        \"ddof\",\n        ddof\n      );\n\n    const xd = toDenseFlatArray(x);\n    let meanX = 0;\n    for (let i = 0; i < n; i++) meanX += xd[i] ?? 0;\n    meanX /= n;\n    let varX = 0;\n    for (let i = 0; i < n; i++) {\n      const dx = (xd[i] ?? 0) - meanX;\n      varX += dx * dx;\n    }\n    varX /= n - ddof;\n    return tensor([[varX]]);\n  }\n\n  if (x.ndim !== 2) {\n    throw new ShapeError(\"cov() expects a 1D or 2D tensor\");\n  }\n\n  const nObs = x.shape[0] ?? 0;\n  const nVar = x.shape[1] ?? 0;\n  if (nObs === 0)\n    throw new InvalidParameterError(\"cov() requires at least one observation\", \"nObs\", nObs);\n  if (ddof < 0) {\n    throw new InvalidParameterError(\"ddof must be non-negative\", \"ddof\", ddof);\n  }\n  if (nObs <= ddof)\n    throw new InvalidParameterError(\n      `ddof=${ddof} >= nObs=${nObs}, covariance undefined`,\n      \"ddof\",\n      ddof\n    );\n\n  const means = new Float64Array(nVar);\n  for (let j = 0; j < nVar; j++) {\n    let s = 0;\n    for (let i = 0; i < nObs; i++) {\n      const off = x.offset + i * (x.strides[0] ?? 0) + j * (x.strides[1] ?? 0);\n      s += getNumberAt(x, off);\n    }\n    means[j] = s / nObs;\n  }\n\n  const out = new Float64Array(nVar * nVar);\n  for (let a = 0; a < nVar; a++) {\n    for (let b = a; b < nVar; b++) {\n      let s = 0;\n      for (let i = 0; i < nObs; i++) {\n        const offA = x.offset + i * (x.strides[0] ?? 0) + a * (x.strides[1] ?? 0);\n        const offB = x.offset + i * (x.strides[0] ?? 0) + b * (x.strides[1] ?? 0);\n        s += (getNumberAt(x, offA) - (means[a] ?? 0)) * (getNumberAt(x, offB) - (means[b] ?? 0));\n      }\n      const v = s / (nObs - ddof);\n      out[a * nVar + b] = v;\n      out[b * nVar + a] = v;\n    }\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: [nVar, nVar],\n    dtype: \"float64\",\n    device: x.device,\n  });\n}\n","import { InvalidParameterError } from \"../core\";\nimport { Tensor, tensor } from \"../ndarray\";\nimport {\n  type AxisLike,\n  computeStrides,\n  forEachIndexOffset,\n  getNumberAt,\n  normalizeAxes,\n  reducedShape,\n  reduceMean,\n  reduceVariance,\n} from \"./_internal\";\n\n/**\n * Computes the arithmetic mean along specified axes.\n *\n * The mean is the sum of all values divided by the count.\n * Supports axis-wise reduction with optional dimension preservation.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute the mean (undefined = all axes)\n * @param _keepdims - If true, reduced axes are retained with size 1 (default: false)\n * @returns Tensor containing mean values\n * @throws {InvalidParameterError} If tensor is empty or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([[1, 2, 3], [4, 5, 6]]);\n * mean(t);           // Returns tensor([3.5]) - mean of all elements\n * mean(t, 0);        // Returns tensor([2.5, 3.5, 4.5]) - column means\n * mean(t, 1);        // Returns tensor([2, 5]) - row means\n * mean(t, 1, true);  // Returns tensor([[2], [5]]) - keepdims\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN output\n * - Infinity is handled according to standard arithmetic rules\n * - Mixed Infinity values (Infinity + -Infinity) result in NaN\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.mean.html | NumPy mean}\n */\nexport function mean(t: Tensor, axis?: AxisLike, _keepdims = false): Tensor {\n  return reduceMean(t, axis, _keepdims);\n}\n\n/**\n * Computes the median (50th percentile) along specified axes.\n *\n * The median is the middle value when data is sorted. For even-sized arrays,\n * it's the average of the two middle values. More robust to outliers than mean.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute the median (undefined = all axes)\n * @param _keepdims - If true, reduced axes are retained with size 1 (default: false)\n * @returns Tensor containing median values\n * @throws {InvalidParameterError} If tensor is empty or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * median(t);  // Returns tensor([3])\n *\n * const t2 = tensor([1, 2, 3, 4]);\n * median(t2); // Returns tensor([2.5]) - average of 2 and 3\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs result in NaN output (NaN sorts to end)\n * - Infinity values are sorted naturally (±Infinity at extremes)\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.median.html | NumPy median}\n */\nexport function median(t: Tensor, axis?: AxisLike, _keepdims = false): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n\n  if (axes.length === 0) {\n    if (t.size === 0) {\n      throw new InvalidParameterError(\"median() requires at least one element\", \"size\", t.size);\n    }\n    const values: number[] = [];\n    let hasNaN = false;\n    forEachIndexOffset(t, (off) => {\n      const v = getNumberAt(t, off);\n      if (Number.isNaN(v)) hasNaN = true;\n      values.push(v);\n    });\n    const outShape = _keepdims ? new Array<number>(t.ndim).fill(1) : [];\n    if (hasNaN) {\n      const out = new Float64Array(1);\n      out[0] = Number.NaN;\n      return Tensor.fromTypedArray({\n        data: out,\n        shape: outShape,\n        dtype: \"float64\",\n        device: t.device,\n      });\n    }\n    values.sort((a, b) => a - b);\n    const mid = Math.floor(values.length / 2);\n    const result =\n      values.length % 2 === 0\n        ? ((values[mid - 1] ?? 0) + (values[mid] ?? 0)) / 2\n        : (values[mid] ?? 0);\n    const out = new Float64Array(1);\n    out[0] = result;\n    return Tensor.fromTypedArray({\n      data: out,\n      shape: outShape,\n      dtype: \"float64\",\n      device: t.device,\n    });\n  }\n\n  const outShape = reducedShape(t.shape, axes, _keepdims);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const buckets: (number[] | undefined)[] = new Array(outSize);\n  const nanFlags = new Array<boolean>(outSize).fill(false);\n\n  const reduce = new Set(axes);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    if (_keepdims) {\n      for (let i = 0; i < t.ndim; i++) {\n        const s = outStrides[i] ?? 0;\n        const v = reduce.has(i) ? 0 : (idx[i] ?? 0);\n        outFlat += v * s;\n      }\n    } else {\n      let oi = 0;\n      for (let i = 0; i < t.ndim; i++) {\n        if (reduce.has(i)) continue;\n        outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n        oi++;\n      }\n    }\n\n    const val = getNumberAt(t, off);\n    if (Number.isNaN(val)) {\n      nanFlags[outFlat] = true;\n      return;\n    }\n    const arr = buckets[outFlat] ?? [];\n    arr.push(val);\n    buckets[outFlat] = arr;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    if (nanFlags[i]) {\n      out[i] = Number.NaN;\n      continue;\n    }\n    const arr = buckets[i] ?? [];\n    if (arr.length === 0) {\n      throw new InvalidParameterError(\n        \"median() reduction over empty axis is undefined\",\n        \"axis\",\n        arr.length\n      );\n    }\n    arr.sort((a, b) => a - b);\n    const mid = Math.floor(arr.length / 2);\n    out[i] = arr.length % 2 === 0 ? ((arr[mid - 1] ?? 0) + (arr[mid] ?? 0)) / 2 : (arr[mid] ?? 0);\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the mode (most frequent value) along specified axis.\n *\n * The mode is the value that appears most frequently in the dataset.\n * If multiple values have the same maximum frequency, returns the smallest value.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute the mode (undefined = all axes)\n * @returns Tensor containing mode values\n * @throws {InvalidParameterError} If tensor is empty\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 2, 3, 3, 3]);\n * mode(t);  // Returns tensor([3]) - most frequent value\n *\n * const t2 = tensor([[1, 2, 2], [3, 3, 4]]);\n * mode(t2, 1);  // Returns tensor([2, 3]) - mode of each row\n * ```\n *\n * @remarks\n * NaN inputs propagate to NaN output.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.mode.html | SciPy stats.mode}\n */\nexport function mode(t: Tensor, axis?: AxisLike): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n\n  if (axes.length === 0) {\n    if (t.size === 0) {\n      throw new InvalidParameterError(\"mode() requires at least one element\", \"size\", t.size);\n    }\n    const freq = new Map<number, number>();\n    let maxFreq = 0;\n    let modeVal = Number.POSITIVE_INFINITY;\n    let hasNaN = false;\n    forEachIndexOffset(t, (off) => {\n      const val = getNumberAt(t, off);\n      if (Number.isNaN(val)) {\n        hasNaN = true;\n        return;\n      }\n      const count = (freq.get(val) ?? 0) + 1;\n      freq.set(val, count);\n      if (count > maxFreq || (count === maxFreq && val < modeVal)) {\n        maxFreq = count;\n        modeVal = val;\n      }\n    });\n    if (hasNaN) {\n      return tensor([Number.NaN]);\n    }\n    return tensor([modeVal]);\n  }\n\n  // Axis-wise mode implemented via per-output frequency maps.\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const maps: Array<Map<number, number> | undefined> = new Array(outSize);\n  const bestCounts = new Int32Array(outSize);\n  const bestValues = new Float64Array(outSize);\n  const nanFlags = new Array<boolean>(outSize).fill(false);\n  const reduce = new Set(axes);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n\n    const val = getNumberAt(t, off);\n    if (Number.isNaN(val)) {\n      nanFlags[outFlat] = true;\n      return;\n    }\n    const m = maps[outFlat] ?? new Map<number, number>();\n    const next = (m.get(val) ?? 0) + 1;\n    m.set(val, next);\n    maps[outFlat] = m;\n    const currentBestCount = bestCounts[outFlat] ?? 0;\n    const currentBestValue = bestValues[outFlat] ?? Number.POSITIVE_INFINITY;\n    if (next > currentBestCount || (next === currentBestCount && val < currentBestValue)) {\n      bestCounts[outFlat] = next;\n      bestValues[outFlat] = val;\n    }\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    if (nanFlags[i]) {\n      out[i] = Number.NaN;\n      continue;\n    }\n    if ((bestCounts[i] ?? 0) === 0) {\n      throw new InvalidParameterError(\n        \"mode() reduction over empty axis is undefined\",\n        \"axis\",\n        bestCounts[i] ?? 0\n      );\n    }\n    out[i] = bestValues[i] ?? Number.NaN;\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the standard deviation along specified axes.\n *\n * Standard deviation is the square root of variance, measuring spread of data.\n * Uses Welford's algorithm for numerical stability via the variance function.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute std (undefined = all axes)\n * @param _keepdims - If true, reduced axes are retained with size 1 (default: false)\n * @param ddof - Delta degrees of freedom (0 = population, 1 = sample, default: 0)\n * @returns Tensor containing standard deviation values\n * @throws {InvalidParameterError} If tensor is empty, ddof < 0, ddof >= sample size, or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n * @throws {DTypeError} If tensor has string dtype\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * std(t);        // Population std (ddof=0)\n * std(t, 0, false, 1);  // Sample std (ddof=1)\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN output\n * - Infinity inputs result in NaN (infinite standard deviation)\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.std.html | NumPy std}\n */\nexport function std(t: Tensor, axis?: AxisLike, _keepdims = false, ddof = 0): Tensor {\n  const v = reduceVariance(t, axis, _keepdims, ddof);\n  const out = new Float64Array(v.size);\n  for (let i = 0; i < v.size; i++) {\n    out[i] = Math.sqrt(getNumberAt(v, v.offset + i));\n  }\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: v.shape,\n    dtype: \"float64\",\n    device: v.device,\n  });\n}\n\n/**\n * Computes the variance along specified axes.\n *\n * Variance measures the average squared deviation from the mean.\n * Uses Welford's online algorithm for numerical stability.\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute variance (undefined = all axes)\n * @param _keepdims - If true, reduced axes are retained with size 1 (default: false)\n * @param ddof - Delta degrees of freedom (0 = population, 1 = sample, default: 0)\n * @returns Tensor containing variance values\n * @throws {InvalidParameterError} If tensor is empty, ddof < 0, ddof >= sample size, or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n * @throws {DTypeError} If tensor has string dtype\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * variance(t);              // Population variance: 2.0\n * variance(t, 0, false, 1); // Sample variance: 2.5\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN output\n * - Infinity inputs result in NaN (infinite variance)\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.var.html | NumPy var}\n */\nexport function variance(t: Tensor, axis?: AxisLike, _keepdims = false, ddof = 0): Tensor {\n  return reduceVariance(t, axis, _keepdims, ddof);\n}\n\n/**\n * Computes the skewness (third standardized moment) along specified axis.\n *\n * Skewness measures the asymmetry of the probability distribution.\n * - Negative skew: left tail is longer (mean < median)\n * - Zero skew: symmetric distribution (normal distribution)\n * - Positive skew: right tail is longer (mean > median)\n *\n * Uses Fisher's moment coefficient: E[(X - μ)³] / σ³\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute skewness (undefined = all axes)\n * @param bias - If false, applies the unbiased Fisher-Pearson correction (default: true)\n * @returns Tensor containing skewness values\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * skewness(t);  // Returns ~0 (symmetric)\n *\n * const t2 = tensor([1, 2, 2, 3, 3, 3, 4, 4, 4, 4]);\n * skewness(t2); // Positive skew (right-tailed)\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN output\n * - Returns NaN for constant input (zero variance)\n * - Unbiased correction requires at least 3 samples; otherwise returns NaN\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.skew.html | SciPy stats.skew}\n */\nexport function skewness(t: Tensor, axis?: AxisLike, bias = true): Tensor {\n  const mu = reduceMean(t, axis, false);\n  const sigma2 = reduceVariance(t, axis, false, 0);\n\n  const axes = normalizeAxes(axis, t.ndim);\n  const reduce = new Set(axes);\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sumCube = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n\n    const m = getNumberAt(mu, mu.offset + outFlat);\n    const v = Math.sqrt(getNumberAt(sigma2, sigma2.offset + outFlat));\n    const x = getNumberAt(t, off);\n    if (!Number.isFinite(v) || v === 0) {\n      sumCube[outFlat] = NaN;\n    } else {\n      const z = (x - m) / v;\n      sumCube[outFlat] = (sumCube[outFlat] ?? 0) + z * z * z;\n    }\n    counts[outFlat] = (counts[outFlat] ?? 0) + 1;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const n = counts[i] ?? 0;\n    if (n === 0) {\n      out[i] = NaN;\n      continue;\n    }\n    let g1 = (sumCube[i] ?? NaN) / n;\n    if (!bias) {\n      // Fisher-Pearson unbiased correction for sample skewness\n      if (n < 3) {\n        g1 = NaN;\n      } else {\n        g1 *= Math.sqrt(n * (n - 1)) / (n - 2);\n      }\n    }\n    out[i] = g1;\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the kurtosis (fourth standardized moment) along specified axis.\n *\n * Kurtosis measures the \"tailedness\" of the probability distribution.\n * - Negative excess kurtosis: lighter tails than normal (platykurtic)\n * - Zero excess kurtosis: same tails as normal distribution (mesokurtic)\n * - Positive excess kurtosis: heavier tails than normal (leptokurtic)\n *\n * Uses Fisher's definition: E[(X - μ)⁴] / σ⁴ - 3 (excess kurtosis)\n *\n * @param t - Input tensor\n * @param axis - Axis or axes along which to compute kurtosis (undefined = all axes)\n * @param fisher - If true, returns excess kurtosis (subtract 3, default: true)\n * @param bias - If false, applies bias correction (requires at least 4 samples, default: true)\n * @returns Tensor containing kurtosis values\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * kurtosis(t, undefined, true);  // Excess kurtosis (Fisher)\n * kurtosis(t, undefined, false); // Raw kurtosis (Pearson)\n * ```\n *\n * @remarks\n * This function follows IEEE 754 semantics for special values:\n * - NaN inputs propagate to NaN output\n * - Returns NaN for constant input (zero variance)\n * - Unbiased correction requires at least 4 samples; otherwise returns NaN\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kurtosis.html | SciPy stats.kurtosis}\n */\nexport function kurtosis(t: Tensor, axis?: AxisLike, fisher = true, bias = true): Tensor {\n  const mu = reduceMean(t, axis, false);\n  const sigma2 = reduceVariance(t, axis, false, 0);\n\n  const axes = normalizeAxes(axis, t.ndim);\n  const reduce = new Set(axes);\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sumQuad = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n\n    const m = getNumberAt(mu, mu.offset + outFlat);\n    const v = Math.sqrt(getNumberAt(sigma2, sigma2.offset + outFlat));\n    const x = getNumberAt(t, off);\n    if (!Number.isFinite(v) || v === 0) {\n      sumQuad[outFlat] = NaN;\n    } else {\n      const z = (x - m) / v;\n      sumQuad[outFlat] = (sumQuad[outFlat] ?? 0) + z ** 4;\n    }\n    counts[outFlat] = (counts[outFlat] ?? 0) + 1;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const n = counts[i] ?? 0;\n    if (n === 0) {\n      out[i] = NaN;\n      continue;\n    }\n    let g2 = (sumQuad[i] ?? NaN) / n; // raw kurtosis\n    if (!bias) {\n      // Unbiased k-statistics correction for excess kurtosis\n      if (n < 4) {\n        g2 = NaN;\n      } else {\n        const excess = g2 - 3;\n        const adj = ((n + 1) * excess + 6) * ((n - 1) / ((n - 2) * (n - 3)));\n        g2 = adj + 3;\n      }\n    }\n    out[i] = fisher ? g2 - 3 : g2;\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes quantiles along specified axes.\n *\n * Quantiles are cut points dividing the range of a probability distribution.\n * Uses linear interpolation between data points.\n *\n * @param t - Input tensor\n * @param q - Quantile(s) to compute, in range [0, 1] (0.5 = median)\n * @param axis - Axis or axes along which to compute quantiles (undefined = all axes)\n * @returns Tensor containing quantile values\n * @throws {InvalidParameterError} If q is not in [0, 1], tensor is empty, or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * quantile(t, 0.5);        // Returns tensor([3]) - median\n * quantile(t, [0.25, 0.75]); // Returns tensor([2, 4]) - quartiles\n * quantile(t, 0.95);       // Returns tensor([4.8]) - 95th percentile\n * ```\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.quantile.html | NumPy quantile}\n */\nexport function quantile(t: Tensor, q: number | number[], axis?: AxisLike): Tensor {\n  const qVals = Array.isArray(q) ? q : [q];\n  for (const v of qVals) {\n    if (!Number.isFinite(v) || v < 0 || v > 1) {\n      throw new InvalidParameterError(\"q must be in [0, 1]\", \"q\", v);\n    }\n  }\n\n  const axes = normalizeAxes(axis, t.ndim);\n  const reduce = new Set(axes);\n\n  if (axes.length === 0) {\n    if (t.size === 0) {\n      throw new InvalidParameterError(\"quantile() requires at least one element\", \"size\", t.size);\n    }\n    const arr: number[] = [];\n    let hasNaN = false;\n    forEachIndexOffset(t, (off) => {\n      const v = getNumberAt(t, off);\n      if (Number.isNaN(v)) hasNaN = true;\n      arr.push(v);\n    });\n    if (hasNaN) {\n      return tensor(qVals.map(() => Number.NaN));\n    }\n    arr.sort((a, b) => a - b);\n    const results: number[] = [];\n    for (const qVal of qVals) {\n      const idx = qVal * (arr.length - 1);\n      const lower = Math.floor(idx);\n      const upper = Math.ceil(idx);\n      const weight = idx - lower;\n      results.push((arr[lower] ?? 0) * (1 - weight) + (arr[upper] ?? 0) * weight);\n    }\n    return tensor(results);\n  }\n\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const buckets: (number[] | undefined)[] = new Array(outSize);\n  const nanFlags = new Array<boolean>(outSize).fill(false);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n    const val = getNumberAt(t, off);\n    if (Number.isNaN(val)) {\n      nanFlags[outFlat] = true;\n      return;\n    }\n    const arr = buckets[outFlat] ?? [];\n    arr.push(val);\n    buckets[outFlat] = arr;\n  });\n\n  // Output shape: (qVals.length, ...reducedShape)\n  const finalShape = [qVals.length, ...outShape];\n  const finalSize = qVals.length * outSize;\n  const out = new Float64Array(finalSize);\n\n  for (let g = 0; g < outSize; g++) {\n    if (nanFlags[g]) {\n      for (let qi = 0; qi < qVals.length; qi++) {\n        out[qi * outSize + g] = Number.NaN;\n      }\n      continue;\n    }\n    const arr = buckets[g] ?? [];\n    if (arr.length === 0) {\n      throw new InvalidParameterError(\n        \"quantile() reduction over empty axis is undefined\",\n        \"axis\",\n        arr.length\n      );\n    }\n    arr.sort((a, b) => a - b);\n    for (let qi = 0; qi < qVals.length; qi++) {\n      const qVal = qVals[qi] ?? 0;\n      const idx = qVal * (arr.length - 1);\n      const lower = Math.floor(idx);\n      const upper = Math.ceil(idx);\n      const weight = idx - lower;\n      out[qi * outSize + g] = (arr[lower] ?? 0) * (1 - weight) + (arr[upper] ?? 0) * weight;\n    }\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: finalShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes percentiles along specified axes.\n *\n * Percentiles are quantiles expressed as percentages (0-100 instead of 0-1).\n * This is a convenience wrapper around quantile().\n *\n * @param t - Input tensor\n * @param q - Percentile(s) to compute, in range [0, 100] (50 = median)\n * @param axis - Axis or axes along which to compute percentiles (undefined = all axes)\n * @returns Tensor containing percentile values\n * @throws {InvalidParameterError} If q is not in [0, 100], tensor is empty, or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * percentile(t, 50);       // Returns tensor([3]) - median\n * percentile(t, [25, 75]); // Returns tensor([2, 4]) - quartiles\n * percentile(t, 95);       // Returns tensor([4.8]) - 95th percentile\n * ```\n *\n * @see {@link https://numpy.org/doc/stable/reference/generated/numpy.percentile.html | NumPy percentile}\n */\nexport function percentile(t: Tensor, q: number | number[], axis?: AxisLike): Tensor {\n  const qArr = Array.isArray(q) ? q : [q];\n  for (const v of qArr) {\n    if (!Number.isFinite(v) || v < 0 || v > 100) {\n      throw new InvalidParameterError(\"q must be in [0, 100]\", \"q\", v);\n    }\n  }\n  // Percentile is just quantile with q/100 (convert percentage to fraction)\n  const qVals = Array.isArray(q) ? q.map((v) => v / 100) : q / 100;\n  return quantile(t, qVals, axis);\n}\n\n/**\n * Computes the n-th central moment about the mean.\n *\n * The n-th moment is defined as: E[(X - μ)ⁿ]\n * - n=1: Always 0 (by definition of mean)\n * - n=2: Variance\n * - n=3: Related to skewness\n * - n=4: Related to kurtosis\n *\n * @param t - Input tensor\n * @param n - Order of the moment (must be non-negative integer)\n * @param axis - Axis or axes along which to compute moment (undefined = all axes)\n * @returns Tensor containing moment values\n * @throws {InvalidParameterError} If n is not a non-negative integer\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5]);\n * moment(t, 1);  // Returns ~0 (first moment about mean)\n * moment(t, 2);  // Returns variance\n * moment(t, 3);  // Returns third moment (related to skewness)\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.moment.html | SciPy stats.moment}\n */\nexport function moment(t: Tensor, n: number, axis?: AxisLike): Tensor {\n  if (!Number.isFinite(n) || !Number.isInteger(n) || n < 0) {\n    throw new InvalidParameterError(\"n must be a non-negative integer\", \"n\", n);\n  }\n\n  const mu = reduceMean(t, axis, false);\n  const axes = normalizeAxes(axis, t.ndim);\n  const reduce = new Set(axes);\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sums = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n    const m = getNumberAt(mu, mu.offset + outFlat);\n    const x = getNumberAt(t, off);\n    sums[outFlat] = (sums[outFlat] ?? 0) + (x - m) ** n;\n    counts[outFlat] = (counts[outFlat] ?? 0) + 1;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const c = counts[i] ?? 0;\n    out[i] = c === 0 ? NaN : (sums[i] ?? 0) / c;\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the geometric mean along specified axis.\n *\n * The geometric mean is the n-th root of the product of n values.\n * Computed as: exp(mean(log(x))) for numerical stability.\n * Useful for averaging ratios, growth rates, and multiplicative processes.\n *\n * @param t - Input tensor (all values must be > 0)\n * @param axis - Axis or axes along which to compute geometric mean (undefined = all axes)\n * @returns Tensor containing geometric mean values\n * @throws {InvalidParameterError} If any value is <= 0\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 4, 8]);\n * geometricMean(t);  // Returns ~2.83 (⁴√(1*2*4*8))\n *\n * // Growth rates: 10% and 20% growth\n * const growth = tensor([1.1, 1.2]);\n * geometricMean(growth);  // Returns ~1.149 (average growth rate)\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.gmean.html | SciPy stats.gmean}\n */\nexport function geometricMean(t: Tensor, axis?: AxisLike): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n  if (axes.length === 0 && t.size === 0) {\n    throw new InvalidParameterError(\n      \"geometricMean() requires at least one element\",\n      \"size\",\n      t.size\n    );\n  }\n  const reduce = new Set(axes);\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sums = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n    const x = getNumberAt(t, off);\n    if (x <= 0)\n      throw new InvalidParameterError(\"geometricMean() requires all values to be > 0\", \"value\", x);\n    sums[outFlat] = (sums[outFlat] ?? 0) + Math.log(x);\n    counts[outFlat] = (counts[outFlat] ?? 0) + 1;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const c = counts[i] ?? 0;\n    if (c === 0) {\n      throw new InvalidParameterError(\n        \"geometricMean() reduction over empty axis is undefined\",\n        \"axis\",\n        c\n      );\n    }\n    out[i] = Math.exp((sums[i] ?? 0) / c);\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the harmonic mean along specified axis.\n *\n * The harmonic mean is the reciprocal of the arithmetic mean of reciprocals.\n * Computed as: n / sum(1/x)\n * Useful for averaging rates and ratios (e.g., speeds, densities).\n *\n * @param t - Input tensor (all values must be > 0)\n * @param axis - Axis or axes along which to compute harmonic mean (undefined = all axes)\n * @returns Tensor containing harmonic mean values\n * @throws {InvalidParameterError} If any value is <= 0\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 4]);\n * harmonicMean(t);  // Returns ~1.71 (3 / (1/1 + 1/2 + 1/4))\n *\n * // Average speed: 60 mph for half distance, 40 mph for other half\n * const speeds = tensor([60, 40]);\n * harmonicMean(speeds);  // Returns 48 mph (correct average)\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.hmean.html | SciPy stats.hmean}\n */\nexport function harmonicMean(t: Tensor, axis?: AxisLike): Tensor {\n  const axes = normalizeAxes(axis, t.ndim);\n  if (axes.length === 0 && t.size === 0) {\n    throw new InvalidParameterError(\"harmonicMean() requires at least one element\", \"size\", t.size);\n  }\n  const reduce = new Set(axes);\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const sums = new Float64Array(outSize);\n  const counts = new Int32Array(outSize);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n    const x = getNumberAt(t, off);\n    if (x <= 0) {\n      throw new InvalidParameterError(\"harmonicMean() requires all values to be > 0\", \"value\", x);\n    }\n    sums[outFlat] = (sums[outFlat] ?? 0) + 1 / x;\n    counts[outFlat] = (counts[outFlat] ?? 0) + 1;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    const c = counts[i] ?? 0;\n    if (c === 0) {\n      throw new InvalidParameterError(\n        \"harmonicMean() reduction over empty axis is undefined\",\n        \"axis\",\n        c\n      );\n    }\n    out[i] = c / (sums[i] ?? NaN);\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n\n/**\n * Computes the trimmed mean (mean after removing outliers from both tails).\n *\n * Removes a specified proportion of extreme values from both ends before computing mean.\n * More robust to outliers than regular mean, less extreme than median.\n *\n * @param t - Input tensor\n * @param proportiontocut - Fraction to cut from each tail, in range [0, 0.5)\n * @param axis - Axis or axes along which to compute trimmed mean (undefined = all axes)\n * @returns Tensor containing trimmed mean values\n * @throws {InvalidParameterError} If proportiontocut is not in [0, 0.5), tensor is empty, or reduction over empty axis\n * @throws {IndexError} If axis is out of bounds\n *\n * @example\n * ```ts\n * const t = tensor([1, 2, 3, 4, 5, 100]); // 100 is outlier\n * mean(t);                    // Returns ~19.17 (affected by outlier)\n * trimMean(t, 0.2);          // Returns 3.5 (removes 1 and 100)\n * trimMean(t, 0.1);          // Returns ~22.8 (removes only 100)\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.trim_mean.html | SciPy stats.trim_mean}\n */\nexport function trimMean(t: Tensor, proportiontocut: number, axis?: AxisLike): Tensor {\n  if (!Number.isFinite(proportiontocut) || proportiontocut < 0 || proportiontocut >= 0.5) {\n    throw new InvalidParameterError(\n      \"proportiontocut must be a finite number in range [0, 0.5)\",\n      \"proportiontocut\",\n      proportiontocut\n    );\n  }\n\n  const axes = normalizeAxes(axis, t.ndim);\n  const reduce = new Set(axes);\n\n  if (axes.length === 0) {\n    const arr: number[] = [];\n    let hasNaN = false;\n    forEachIndexOffset(t, (off) => {\n      const v = getNumberAt(t, off);\n      if (Number.isNaN(v)) hasNaN = true;\n      arr.push(v);\n    });\n    if (arr.length === 0)\n      throw new InvalidParameterError(\n        \"trimMean() requires at least one element\",\n        \"size\",\n        arr.length\n      );\n    if (hasNaN) {\n      return tensor([Number.NaN]);\n    }\n    arr.sort((a, b) => a - b);\n    const nTrim = Math.floor(arr.length * proportiontocut);\n    const trimmed = arr.slice(nTrim, arr.length - nTrim);\n    const sum = trimmed.reduce((a, b) => a + b, 0);\n    return tensor([sum / trimmed.length]);\n  }\n\n  const outShape = reducedShape(t.shape, axes, false);\n  const outStrides = computeStrides(outShape);\n  const outSize = outShape.reduce((a, b) => a * b, 1);\n  const buckets: (number[] | undefined)[] = new Array(outSize);\n  const nanFlags = new Array<boolean>(outSize).fill(false);\n\n  forEachIndexOffset(t, (off, idx) => {\n    let outFlat = 0;\n    let oi = 0;\n    for (let i = 0; i < t.ndim; i++) {\n      if (reduce.has(i)) continue;\n      outFlat += (idx[i] ?? 0) * (outStrides[oi] ?? 0);\n      oi++;\n    }\n    const val = getNumberAt(t, off);\n    if (Number.isNaN(val)) {\n      nanFlags[outFlat] = true;\n      return;\n    }\n    const arr = buckets[outFlat] ?? [];\n    arr.push(val);\n    buckets[outFlat] = arr;\n  });\n\n  const out = new Float64Array(outSize);\n  for (let i = 0; i < outSize; i++) {\n    if (nanFlags[i]) {\n      out[i] = Number.NaN;\n      continue;\n    }\n    const arr = buckets[i] ?? [];\n    if (arr.length === 0) {\n      throw new InvalidParameterError(\n        \"trimMean() reduction over empty axis is undefined\",\n        \"axis\",\n        arr.length\n      );\n    }\n    arr.sort((a, b) => a - b);\n    const nTrim = Math.floor(arr.length * proportiontocut);\n    const trimmed = arr.slice(nTrim, arr.length - nTrim);\n    const sum = trimmed.reduce((a, b) => a + b, 0);\n    out[i] = sum / trimmed.length;\n  }\n\n  return Tensor.fromTypedArray({\n    data: out,\n    shape: outShape,\n    dtype: \"float64\",\n    device: t.device,\n  });\n}\n","import { InvalidParameterError } from \"../core\";\nimport type { Tensor } from \"../ndarray\";\nimport {\n  chiSquareCdf,\n  fCdf,\n  forEachIndexOffset,\n  getNumberAt,\n  normalCdf,\n  rankData,\n  studentTCdf,\n} from \"./_internal\";\n\nexport type TestResult = {\n  statistic: number;\n  pvalue: number;\n};\n\nfunction toDenseSortedArray1D(x: Tensor): Float64Array {\n  if (x.size < 1) {\n    return new Float64Array(0);\n  }\n  const out = new Float64Array(x.size);\n  let i = 0;\n  forEachIndexOffset(x, (off) => {\n    out[i] = getNumberAt(x, off);\n    i++;\n  });\n  out.sort((a, b) => a - b);\n  return out;\n}\n\nfunction toDenseArray1D(x: Tensor): Float64Array {\n  const out = new Float64Array(x.size);\n  let i = 0;\n  forEachIndexOffset(x, (off) => {\n    out[i] = getNumberAt(x, off);\n    i++;\n  });\n  return out;\n}\n\nfunction meanAndM2(x: Float64Array): { mean: number; m2: number } {\n  if (x.length === 0) {\n    throw new InvalidParameterError(\"expected at least one element\", \"length\", x.length);\n  }\n  let mean = 0;\n  let m2 = 0;\n  for (let i = 0; i < x.length; i++) {\n    const v = x[i] ?? 0;\n    const n = i + 1;\n    const delta = v - mean;\n    mean += delta / n;\n    const delta2 = v - mean;\n    m2 += delta * delta2;\n  }\n  return { mean, m2 };\n}\n\nfunction shapiroWilk(x: Float64Array): TestResult {\n  // Ported from Algorithm AS R94 (Royston, 1995), following implementations\n  // derived from the original FORTRAN and common ports.\n  const n = x.length;\n  if (n < 3 || n > 5000) {\n    throw new InvalidParameterError(\"shapiro() sample size must be between 3 and 5000\", \"n\", n);\n  }\n\n  const range = (x[n - 1] ?? 0) - (x[0] ?? 0);\n  if (range === 0) {\n    throw new InvalidParameterError(\"shapiro() all x values are identical\", \"range\", range);\n  }\n\n  const small = 1e-19;\n  if (range < small) {\n    throw new InvalidParameterError(\"shapiro() range is too small\", \"range\", range);\n  }\n\n  const nn2 = Math.floor(n / 2);\n  const a = new Float64Array(nn2 + 1); // 1-based\n\n  const g: readonly number[] = [-2.273, 0.459];\n  const c1: readonly number[] = [0.0, 0.221157, -0.147981, -2.07119, 4.434685, -2.706056];\n  const c2: readonly number[] = [0.0, 0.042981, -0.293762, -1.752461, 5.682633, -3.582633];\n  const c3: readonly number[] = [0.544, -0.39978, 0.025054, -6.714e-4];\n  const c4: readonly number[] = [1.3822, -0.77857, 0.062767, -0.0020322];\n  const c5: readonly number[] = [-1.5861, -0.31082, -0.083751, 0.0038915];\n  const c6: readonly number[] = [-0.4803, -0.082676, 0.0030302];\n\n  const poly = (cc: readonly number[], x0: number): number => {\n    let p = cc[cc.length - 1] ?? 0;\n    for (let j = cc.length - 2; j >= 0; j--) {\n      p = p * x0 + (cc[j] ?? 0);\n    }\n    return p;\n  };\n\n  const sign = (v: number): number => (v === 0 ? 0 : v > 0 ? 1 : -1);\n\n  const an = n;\n  if (n === 3) {\n    a[1] = Math.SQRT1_2;\n  } else {\n    const an25 = an + 0.25;\n    let summ2 = 0;\n    for (let i = 1; i <= nn2; i++) {\n      const p = (i - 0.375) / an25;\n      // Inverse normal CDF via approximation using studentTCdf? We use a rational approximation:\n      // For this package, approximate quantile using inverse error function via binary search on normalCdf.\n      let lo = -10;\n      let hi = 10;\n      for (let it = 0; it < 80; it++) {\n        const mid = (lo + hi) / 2;\n        const cdf = normalCdf(mid);\n        if (cdf < p) lo = mid;\n        else hi = mid;\n      }\n      const z = (lo + hi) / 2;\n      a[i] = z;\n      summ2 += z * z;\n    }\n    summ2 *= 2;\n    const ssumm2 = Math.sqrt(summ2);\n    const rsn = 1 / Math.sqrt(an);\n\n    const a1 = poly(c1, rsn) - (a[1] ?? 0) / ssumm2;\n    let i1: number;\n    let fac: number;\n    if (n > 5) {\n      i1 = 3;\n      const a2 = -((a[2] ?? 0) / ssumm2) + poly(c2, rsn);\n      fac = Math.sqrt(\n        (summ2 - 2 * (a[1] ?? 0) * (a[1] ?? 0) - 2 * (a[2] ?? 0) * (a[2] ?? 0)) /\n          (1 - 2 * a1 * a1 - 2 * a2 * a2)\n      );\n      a[2] = a2;\n    } else {\n      i1 = 2;\n      fac = Math.sqrt((summ2 - 2 * (a[1] ?? 0) * (a[1] ?? 0)) / (1 - 2 * a1 * a1));\n    }\n    a[1] = a1;\n    for (let i = i1; i <= nn2; i++) {\n      a[i] = -((a[i] ?? 0) / fac);\n    }\n  }\n\n  // Check sort order and compute scaled sums\n  let xx = (x[0] ?? 0) / range;\n  let sx = xx;\n  let sa = -(a[1] ?? 0);\n  for (let i = 1, j = n - 1; i < n; j--) {\n    const xi = (x[i] ?? 0) / range;\n    if (xx - xi > small) {\n      throw new InvalidParameterError(\"shapiro() data is not sorted\", \"data\", \"unsorted\");\n    }\n    sx += xi;\n    i++;\n    if (i !== j) {\n      sa += sign(i - j) * (a[Math.min(i, j)] ?? 0);\n    }\n    xx = xi;\n  }\n\n  sa /= n;\n  sx /= n;\n\n  let ssa = 0;\n  let ssx = 0;\n  let sax = 0;\n  for (let i = 0, j = n - 1; i < n; i++, j--) {\n    const asa = i !== j ? sign(i - j) * (a[1 + Math.min(i, j)] ?? 0) - sa : -sa;\n    const xsx = (x[i] ?? 0) / range - sx;\n    ssa += asa * asa;\n    ssx += xsx * xsx;\n    sax += asa * xsx;\n  }\n\n  const ssassx = Math.sqrt(ssa * ssx);\n  const w1 = ((ssassx - sax) * (ssassx + sax)) / (ssa * ssx);\n  const w = 1 - w1;\n\n  if (n === 3) {\n    const pi6 = 1.90985931710274;\n    const stqr = 1.0471975511966;\n    let pw = pi6 * (Math.asin(Math.sqrt(w)) - stqr);\n    if (pw < 0) pw = 0;\n    if (pw > 1) pw = 1;\n    return { statistic: w, pvalue: pw };\n  }\n\n  const y = Math.log(w1);\n  const lnN = Math.log(an);\n  let m: number;\n  let s: number;\n  if (n <= 11) {\n    const gamma = poly(g, an);\n    if (y >= gamma) {\n      return { statistic: w, pvalue: 0 };\n    }\n    const yy = -Math.log(gamma - y);\n    m = poly(c3, an);\n    s = Math.exp(poly(c4, an));\n    const z = (yy - m) / s;\n    return { statistic: w, pvalue: normalCdf(z) };\n  }\n  m = poly(c5, lnN);\n  s = Math.exp(poly(c6, lnN));\n  const z = (y - m) / s;\n  return { statistic: w, pvalue: normalCdf(z) };\n}\n\n/**\n * One-sample t-test.\n *\n * Tests whether mean of sample differs from population mean.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_1samp.html | SciPy stats.ttest_1samp}\n */\nexport function ttest_1samp(a: Tensor, popmean: number): TestResult {\n  const x = toDenseSortedArray1D(a);\n  const n = x.length;\n  if (n < 2) {\n    throw new InvalidParameterError(\"ttest_1samp() requires at least 2 samples\", \"n\", n);\n  }\n\n  // mean\n  let mean = 0;\n  for (let i = 0; i < n; i++) mean += x[i] ?? 0;\n  mean /= n;\n\n  // sample variance\n  let m2 = 0;\n  for (let i = 0; i < n; i++) {\n    const d = (x[i] ?? 0) - mean;\n    m2 += d * d;\n  }\n  const variance = m2 / (n - 1);\n  const std = Math.sqrt(variance);\n  if (std === 0) {\n    throw new InvalidParameterError(\"ttest_1samp() is undefined for constant input\", \"std\", std);\n  }\n  const tstat = (mean - popmean) / (std / Math.sqrt(n));\n\n  const df = n - 1;\n  const pvalue = 2 * (1 - studentTCdf(Math.abs(tstat), df));\n  return { statistic: tstat, pvalue };\n}\n\n/**\n * Independent two-sample t-test.\n *\n * Tests whether means of two independent samples are equal.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_ind.html | SciPy stats.ttest_ind}\n */\nexport function ttest_ind(a: Tensor, b: Tensor, equalVar = true): TestResult {\n  const xa = toDenseSortedArray1D(a);\n  const xb = toDenseSortedArray1D(b);\n  const na = xa.length;\n  const nb = xb.length;\n  if (na < 2 || nb < 2) {\n    throw new InvalidParameterError(\"ttest_ind() requires at least 2 samples in each group\", \"n\", {\n      na,\n      nb,\n    });\n  }\n\n  let meanA = 0;\n  let meanB = 0;\n  for (let i = 0; i < na; i++) meanA += xa[i] ?? 0;\n  for (let i = 0; i < nb; i++) meanB += xb[i] ?? 0;\n  meanA /= na;\n  meanB /= nb;\n\n  let ssa = 0;\n  let ssb = 0;\n  for (let i = 0; i < na; i++) {\n    const d = (xa[i] ?? 0) - meanA;\n    ssa += d * d;\n  }\n  for (let i = 0; i < nb; i++) {\n    const d = (xb[i] ?? 0) - meanB;\n    ssb += d * d;\n  }\n  const varA = ssa / (na - 1);\n  const varB = ssb / (nb - 1);\n\n  let tstat: number;\n  let df: number;\n\n  if (equalVar) {\n    const pooledVar = ((na - 1) * varA + (nb - 1) * varB) / (na + nb - 2);\n    const denom = Math.sqrt(pooledVar * (1 / na + 1 / nb));\n    if (denom === 0)\n      throw new InvalidParameterError(\n        \"ttest_ind() is undefined for constant input\",\n        \"denom\",\n        denom\n      );\n    tstat = (meanA - meanB) / denom;\n    df = na + nb - 2;\n  } else {\n    const denom = Math.sqrt(varA / na + varB / nb);\n    if (denom === 0)\n      throw new InvalidParameterError(\n        \"ttest_ind() is undefined for constant input\",\n        \"denom\",\n        denom\n      );\n    tstat = (meanA - meanB) / denom;\n    df = (varA / na + varB / nb) ** 2 / ((varA / na) ** 2 / (na - 1) + (varB / nb) ** 2 / (nb - 1));\n  }\n\n  const pvalue = 2 * (1 - studentTCdf(Math.abs(tstat), df));\n  return { statistic: tstat, pvalue };\n}\n\n/**\n * Paired-sample t-test.\n *\n * Tests whether means of two related samples are equal.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_rel.html | SciPy stats.ttest_rel}\n */\nexport function ttest_rel(a: Tensor, b: Tensor): TestResult {\n  if (a.size !== b.size) {\n    throw new InvalidParameterError(\"ttest_rel() requires paired samples of equal length\", \"size\", {\n      a: a.size,\n      b: b.size,\n    });\n  }\n  const n = a.size;\n  if (n < 2) {\n    throw new InvalidParameterError(\"ttest_rel() requires at least 2 paired samples\", \"n\", n);\n  }\n\n  // Differences\n  const diffs = new Float64Array(n);\n  let i = 0;\n  forEachIndexOffset(a, (offA) => {\n    // Map to corresponding flat element in b using the same iteration order.\n    // We rely on size equality and iterate both tensors to dense arrays.\n    diffs[i] = getNumberAt(a, offA);\n    i++;\n  });\n  const bd = new Float64Array(n);\n  i = 0;\n  forEachIndexOffset(b, (offB) => {\n    bd[i] = getNumberAt(b, offB);\n    i++;\n  });\n  for (let k = 0; k < n; k++) {\n    diffs[k] = (diffs[k] ?? 0) - (bd[k] ?? 0);\n  }\n\n  let mean = 0;\n  for (let k = 0; k < n; k++) mean += diffs[k] ?? 0;\n  mean /= n;\n\n  let ss = 0;\n  for (let k = 0; k < n; k++) {\n    const d = (diffs[k] ?? 0) - mean;\n    ss += d * d;\n  }\n  const varDiff = ss / (n - 1);\n  const stdDiff = Math.sqrt(varDiff);\n  if (stdDiff === 0) {\n    throw new InvalidParameterError(\n      \"ttest_rel() is undefined for constant differences\",\n      \"stdDiff\",\n      stdDiff\n    );\n  }\n  const tstat = mean / (stdDiff / Math.sqrt(n));\n  const df = n - 1;\n  const pvalue = 2 * (1 - studentTCdf(Math.abs(tstat), df));\n  return { statistic: tstat, pvalue };\n}\n\n/**\n * Chi-square goodness of fit test.\n *\n * Observed and expected frequencies must be non-negative and sum to the same total\n * (within floating-point tolerance).\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.chisquare.html | SciPy stats.chisquare}\n */\nexport function chisquare(f_obs: Tensor, f_exp?: Tensor): TestResult {\n  const obs = toDenseArray1D(f_obs);\n  const n = obs.length;\n  if (n < 1) {\n    throw new InvalidParameterError(\"chisquare() requires at least one observed value\", \"n\", n);\n  }\n  let chiSq = 0;\n  let sumObs = 0;\n\n  for (let i = 0; i < n; i++) {\n    const v = obs[i] ?? 0;\n    if (!Number.isFinite(v) || v < 0) {\n      throw new InvalidParameterError(\n        \"chisquare() observed frequencies must be finite and >= 0\",\n        \"f_obs\",\n        v\n      );\n    }\n    sumObs += v;\n  }\n\n  if (f_exp && f_obs.size !== f_exp.size) {\n    throw new InvalidParameterError(\n      \"Observed and expected frequency arrays must have the same length\",\n      \"size\",\n      { f_obs: f_obs.size, f_exp: f_exp.size }\n    );\n  }\n\n  if (!f_exp) {\n    // Uniform expected frequencies\n    const expected = sumObs / n;\n    if (!Number.isFinite(expected) || expected <= 0) {\n      throw new InvalidParameterError(\n        \"chisquare() expected frequencies must be finite and > 0\",\n        \"expected\",\n        expected\n      );\n    }\n    for (let i = 0; i < n; i++) {\n      const v = obs[i] ?? 0;\n      chiSq += (v - expected) ** 2 / expected;\n    }\n  } else {\n    const exp = toDenseArray1D(f_exp);\n    let sumExp = 0;\n    for (let i = 0; i < n; i++) {\n      const v = exp[i] ?? 0;\n      if (!Number.isFinite(v) || v <= 0) {\n        throw new InvalidParameterError(\n          \"chisquare() expected frequencies must be finite and > 0\",\n          \"f_exp\",\n          v\n        );\n      }\n      sumExp += v;\n    }\n\n    const rtol = Math.sqrt(Number.EPSILON);\n    const denom = Math.max(Math.abs(sumObs), Math.abs(sumExp));\n    if (Math.abs(sumObs - sumExp) > rtol * denom) {\n      throw new InvalidParameterError(\n        \"chisquare() expected and observed frequencies must sum to the same value\",\n        \"sum\",\n        { f_obs: sumObs, f_exp: sumExp }\n      );\n    }\n\n    for (let i = 0; i < n; i++) {\n      const vObs = obs[i] ?? 0;\n      const vExp = exp[i] ?? 0;\n      chiSq += (vObs - vExp) ** 2 / vExp;\n    }\n  }\n\n  const df = n - 1;\n  if (df < 1) {\n    throw new InvalidParameterError(\n      \"chisquare() requires at least 2 categories (df must be >= 1)\",\n      \"df\",\n      df\n    );\n  }\n  const pvalue = 1 - chiSquareCdf(chiSq, df);\n\n  return { statistic: chiSq, pvalue };\n}\n\n/**\n * Kolmogorov-Smirnov test for goodness of fit.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kstest.html | SciPy stats.kstest}\n */\nexport function kstest(data: Tensor, cdf: string | ((x: number) => number)): TestResult {\n  const x = toDenseSortedArray1D(data);\n  const n = x.length;\n  if (n === 0) {\n    throw new InvalidParameterError(\"kstest() requires at least one element\", \"n\", n);\n  }\n\n  if (typeof cdf === \"string\" && cdf !== \"norm\") {\n    throw new InvalidParameterError(\n      `Unsupported distribution: '${cdf}'. Supported distributions: 'norm'`,\n      \"cdf\",\n      cdf\n    );\n  }\n\n  const F = typeof cdf === \"string\" ? (v: number) => normalCdf(v) : cdf;\n\n  let d = 0;\n  for (let i = 0; i < n; i++) {\n    const xi = x[i] ?? 0;\n    const fi = F(xi);\n    const dPlus = (i + 1) / n - fi;\n    const dMinus = fi - i / n;\n    d = Math.max(d, dPlus, dMinus);\n  }\n\n  // Asymptotic p-value (two-sided) using Kolmogorov distribution\n  // p ≈ 2 * sum_{k=1..inf} (-1)^{k-1} exp(-2 k^2 d^2 n)\n  let p = 0;\n  for (let k = 1; k < 200; k++) {\n    const term = Math.exp(-2 * k * k * d * d * n);\n    p += (k % 2 === 1 ? 1 : -1) * term;\n    if (term < 1e-12) break;\n  }\n  p = Math.max(0, Math.min(1, 2 * p));\n\n  return { statistic: d, pvalue: p };\n}\n\n/**\n * Test for normality.\n *\n * Uses D'Agostino-Pearson's omnibus test combining skewness and kurtosis.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.normaltest.html | SciPy stats.normaltest}\n */\nexport function normaltest(a: Tensor): TestResult {\n  const x = toDenseSortedArray1D(a);\n  const n = x.length;\n  if (n < 8) {\n    throw new InvalidParameterError(\"normaltest() requires at least 8 samples\", \"n\", n);\n  }\n  const { mean, m2 } = meanAndM2(x);\n  const m2n = m2 / n;\n  const std = Math.sqrt(m2n);\n  if (!Number.isFinite(std) || std === 0) {\n    throw new InvalidParameterError(\"normaltest() is undefined for constant input\", \"std\", std);\n  }\n\n  let m3 = 0;\n  let m4 = 0;\n  for (let i = 0; i < n; i++) {\n    const d = (x[i] ?? 0) - mean;\n    m3 += d * d * d;\n    m4 += d * d * d * d;\n  }\n  const skew = m3 / n / m2n ** 1.5;\n  const kurt = m4 / n / (m2n * m2n);\n\n  // Skewness component (D'Agostino)\n  const y = skew * Math.sqrt(((n + 1) * (n + 3)) / (6 * (n - 2)));\n  const beta2 =\n    (3 * (n * n + 27 * n - 70) * (n + 1) * (n + 3)) / ((n - 2) * (n + 5) * (n + 7) * (n + 9));\n  const w2 = -1 + Math.sqrt(2 * (beta2 - 1));\n  const delta = 1 / Math.sqrt(0.5 * Math.log(w2));\n  const alpha = Math.sqrt(2 / (w2 - 1));\n  const yScaled = y / alpha;\n  const z1 = delta * Math.log(yScaled + Math.sqrt(yScaled * yScaled + 1));\n\n  // Kurtosis component (Anscombe-Glynn)\n  const e = (3 * (n - 1)) / (n + 1);\n  const varb2 = (24 * n * (n - 2) * (n - 3)) / ((n + 1) * (n + 1) * (n + 3) * (n + 5));\n  const xval = (kurt - e) / Math.sqrt(varb2);\n  const sqrtbeta1 =\n    ((6 * (n * n - 5 * n + 2)) / ((n + 7) * (n + 9))) *\n    Math.sqrt((6 * (n + 3) * (n + 5)) / (n * (n - 2) * (n - 3)));\n  const aTerm = 6 + (8 / sqrtbeta1) * (2 / sqrtbeta1 + Math.sqrt(1 + 4 / (sqrtbeta1 * sqrtbeta1)));\n  const term1 = 1 - 2 / (9 * aTerm);\n  const denom = 1 + xval * Math.sqrt(2 / (aTerm - 4));\n  const term2 =\n    denom === 0 ? Number.NaN : Math.sign(denom) * ((1 - 2 / aTerm) / Math.abs(denom)) ** (1 / 3);\n  const z2 = (term1 - term2) / Math.sqrt(2 / (9 * aTerm));\n\n  const k2 = z1 * z1 + z2 * z2;\n  const pvalue = 1 - chiSquareCdf(k2, 2);\n  return { statistic: k2, pvalue };\n}\n\n/**\n * Shapiro-Wilk test for normality.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.shapiro.html | SciPy stats.shapiro}\n */\nexport function shapiro(x: Tensor): TestResult {\n  const sorted = toDenseSortedArray1D(x);\n  return shapiroWilk(sorted);\n}\n\n/**\n * Anderson-Darling test for normality.\n *\n * Uses sample standard deviation and size-adjusted critical values.\n * For very small samples (n < 10), uses an IQR-based scale estimate to stabilize\n * the statistic.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.anderson.html | SciPy stats.anderson}\n */\nexport function anderson(x: Tensor): {\n  statistic: number;\n  critical_values: number[];\n  significance_level: number[];\n} {\n  const sorted = toDenseSortedArray1D(x);\n  const n = sorted.length;\n  if (n < 1) {\n    throw new InvalidParameterError(\"anderson() requires at least one element\", \"n\", n);\n  }\n  const { mean, m2 } = meanAndM2(sorted);\n  const variance = n > 1 ? m2 / (n - 1) : NaN;\n  let std = Math.sqrt(variance);\n  if (!Number.isFinite(std) || std === 0) {\n    throw new InvalidParameterError(\"anderson() is undefined for constant input\", \"std\", std);\n  }\n  if (n < 10) {\n    const quantile = (q: number): number => {\n      if (n === 1) return sorted[0] ?? 0;\n      const pos = (n - 1) * q;\n      const lo = Math.floor(pos);\n      const hi = Math.ceil(pos);\n      const v0 = sorted[lo] ?? 0;\n      const v1 = sorted[hi] ?? v0;\n      return v0 + (pos - lo) * (v1 - v0);\n    };\n    const q1 = quantile(0.25);\n    const q3 = quantile(0.75);\n    const iqr = q3 - q1;\n    const robust = iqr / 1.349;\n    if (Number.isFinite(robust) && robust > 0) {\n      std = robust;\n    }\n  }\n\n  // Anderson-Darling for normal: A^2 = -n - (1/n) sum_{i=1..n} (2i-1)[ln Phi(z_i) + ln(1-Phi(z_{n+1-i}))]\n  let A2 = 0;\n  for (let i = 0; i < n; i++) {\n    const zi = ((sorted[i] ?? 0) - mean) / std;\n    const zj = ((sorted[n - 1 - i] ?? 0) - mean) / std;\n    const PhiI = Math.max(1e-300, Math.min(1 - 1e-16, normalCdf(zi)));\n    const PhiJ = Math.max(1e-300, Math.min(1 - 1e-16, normalCdf(zj)));\n    A2 += (2 * (i + 1) - 1) * (Math.log(PhiI) + Math.log(1 - PhiJ));\n  }\n  A2 = -n - A2 / n;\n\n  const baseCritical = [0.576, 0.656, 0.787, 0.918, 1.092];\n  const factor = 1 + 4 / n - 25 / (n * n);\n  const critical_values = baseCritical.map((v) => Math.round((v / factor) * 1000) / 1000);\n\n  return {\n    statistic: A2,\n    critical_values,\n    significance_level: [0.15, 0.1, 0.05, 0.025, 0.01],\n  };\n}\n\n/**\n * Mann-Whitney U test (non-parametric).\n *\n * Tests whether two independent samples come from same distribution.\n *\n * Note: Uses normal approximation for the p-value with tie correction and\n * continuity correction. No exact method selection is available.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.mannwhitneyu.html | SciPy stats.mannwhitneyu}\n */\nexport function mannwhitneyu(x: Tensor, y: Tensor): TestResult {\n  const nx = x.size;\n  const ny = y.size;\n\n  if (nx < 1 || ny < 1) {\n    throw new InvalidParameterError(\"Both samples must be non-empty\", \"size\", {\n      x: nx,\n      y: ny,\n    });\n  }\n\n  const xVals = toDenseArray1D(x);\n  const yVals = toDenseArray1D(y);\n  const n = nx + ny;\n  const combined = new Float64Array(n);\n  combined.set(xVals, 0);\n  combined.set(yVals, nx);\n\n  const { ranks, tieSum } = rankData(combined);\n\n  // Rank sum for group 1 (first sample)\n  let R1 = 0;\n  for (let i = 0; i < nx; i++) {\n    R1 += ranks[i] ?? 0;\n  }\n\n  const U1 = R1 - (nx * (nx + 1)) / 2;\n  const U2 = nx * ny - U1;\n  const U = Math.min(U1, U2);\n\n  const meanU = (nx * ny) / 2;\n  // Tie correction uses sum(t^3 - t) over tied groups.\n  const tieAdj = n > 1 ? tieSum / (n * (n - 1)) : 0;\n  const varU = (nx * ny * (n + 1 - tieAdj)) / 12;\n  if (varU <= 0) {\n    return { statistic: U, pvalue: NaN };\n  }\n\n  const stdU = Math.sqrt(varU);\n  // Continuity correction for two-sided normal approximation.\n  // For small samples, omit correction to better match exact distribution behavior.\n  const useContinuity = nx + ny > 20;\n  const continuity = useContinuity ? (U < meanU ? 0.5 : U > meanU ? -0.5 : 0) : 0;\n  const z = (U - meanU + continuity) / stdU;\n  const pvalue = 2 * (1 - normalCdf(Math.abs(z)));\n\n  return { statistic: U, pvalue };\n}\n\n/**\n * Wilcoxon signed-rank test (non-parametric paired test).\n *\n * Note: Uses normal approximation for the p-value with tie correction and\n * continuity correction. No exact method selection is available.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.wilcoxon.html | SciPy stats.wilcoxon}\n */\nexport function wilcoxon(x: Tensor, y?: Tensor): TestResult {\n  const n = x.size;\n  const diffs: number[] = [];\n\n  if (y) {\n    if (x.size !== y.size) {\n      throw new InvalidParameterError(\"Paired samples must have equal length\", \"size\", {\n        x: x.size,\n        y: y.size,\n      });\n    }\n    const xd = toDenseArray1D(x);\n    const yd = toDenseArray1D(y);\n    for (let i = 0; i < n; i++) {\n      const diff = (xd[i] ?? 0) - (yd[i] ?? 0);\n      if (diff !== 0) diffs.push(diff);\n    }\n  } else {\n    const xd = toDenseArray1D(x);\n    for (let i = 0; i < n; i++) {\n      const val = xd[i] ?? 0;\n      if (val !== 0) diffs.push(val);\n    }\n  }\n\n  if (diffs.length === 0) {\n    throw new InvalidParameterError(\n      \"wilcoxon() is undefined when all differences are zero\",\n      \"diffs\",\n      diffs.length\n    );\n  }\n\n  const absDiffs = new Float64Array(diffs.length);\n  for (let i = 0; i < diffs.length; i++) {\n    absDiffs[i] = Math.abs(diffs[i] ?? 0);\n  }\n\n  const { ranks, tieSum } = rankData(absDiffs);\n\n  let Wplus = 0;\n  for (let i = 0; i < diffs.length; i++) {\n    if ((diffs[i] ?? 0) > 0) {\n      Wplus += ranks[i] ?? 0;\n    }\n  }\n\n  const nEff = diffs.length;\n  const meanW = (nEff * (nEff + 1)) / 4;\n  // Tie-corrected variance for signed-rank statistic.\n  const varW = (nEff * (nEff + 1) * (2 * nEff + 1)) / 24 - tieSum / 48;\n  if (varW <= 0) {\n    return { statistic: Wplus, pvalue: NaN };\n  }\n\n  const stdW = Math.sqrt(varW);\n  // Continuity correction for two-sided normal approximation.\n  // For small samples, omit correction to better match exact distribution behavior.\n  const useContinuity = nEff > 20;\n  const continuity = useContinuity ? (Wplus < meanW ? 0.5 : Wplus > meanW ? -0.5 : 0) : 0;\n  const z = (Wplus - meanW + continuity) / stdW;\n  const pvalue = 2 * (1 - normalCdf(Math.abs(z)));\n\n  return { statistic: Wplus, pvalue };\n}\n\n/**\n * Kruskal-Wallis H-test (non-parametric version of ANOVA).\n *\n * Note: Uses chi-square approximation for the p-value with tie correction.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.kruskal.html | SciPy stats.kruskal}\n */\nexport function kruskal(...samples: Tensor[]): TestResult {\n  const k = samples.length;\n  if (k < 2) {\n    throw new InvalidParameterError(\"kruskal() requires at least 2 groups\", \"k\", k);\n  }\n\n  let N = 0;\n  const sizes = new Array<number | undefined>(k);\n  const validatedSamples = new Array<Tensor | undefined>(k);\n  for (let g = 0; g < k; g++) {\n    const sample = samples[g];\n    if (!sample || sample.size < 1) {\n      throw new InvalidParameterError(\"kruskal() requires non-empty samples\", \"size\", {\n        group: g,\n        size: sample?.size ?? 0,\n      });\n    }\n    validatedSamples[g] = sample;\n    sizes[g] = sample.size;\n    N += sample.size;\n  }\n\n  const combined = new Float64Array(N);\n  const groupIndex = new Int32Array(N);\n  let idx = 0;\n  for (let g = 0; g < k; g++) {\n    const sample = validatedSamples[g];\n    if (!sample) {\n      throw new InvalidParameterError(\"kruskal() requires non-empty samples\", \"sample\", g);\n    }\n    const vals = toDenseArray1D(sample);\n    for (let i = 0; i < vals.length; i++) {\n      combined[idx] = vals[i] ?? 0;\n      groupIndex[idx] = g;\n      idx++;\n    }\n  }\n\n  const { ranks, tieSum } = rankData(combined);\n\n  const rankSums = new Float64Array(k);\n  for (let i = 0; i < N; i++) {\n    const group = groupIndex[i] ?? 0;\n    const rank = ranks[i] ?? 0;\n    rankSums[group] = (rankSums[group] ?? 0) + rank;\n  }\n\n  let H = 0;\n  for (let g = 0; g < k; g++) {\n    const rs = rankSums[g] ?? 0;\n    const sz = sizes[g] ?? 1;\n    H += (rs * rs) / sz;\n  }\n  H = (12 / (N * (N + 1))) * H - 3 * (N + 1);\n\n  // Tie correction factor for H statistic.\n  const tieCorrection = N > 1 ? 1 - tieSum / (N * N * N - N) : 1;\n  if (tieCorrection <= 0) {\n    throw new InvalidParameterError(\n      \"kruskal() is undefined when all numbers are identical\",\n      \"tieCorrection\",\n      tieCorrection\n    );\n  }\n  H /= tieCorrection;\n\n  const df = k - 1;\n  const pvalue = 1 - chiSquareCdf(H, df);\n\n  return { statistic: H, pvalue };\n}\n\n/**\n * Friedman test (non-parametric repeated measures ANOVA).\n *\n * Note: Uses chi-square approximation for the p-value with tie correction.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.friedmanchisquare.html | SciPy stats.friedmanchisquare}\n */\nexport function friedmanchisquare(...samples: Tensor[]): TestResult {\n  const k = samples.length;\n  if (k < 3) {\n    throw new InvalidParameterError(\n      \"friedmanchisquare() requires at least 3 related samples\",\n      \"k\",\n      k\n    );\n  }\n  const n = samples[0]?.size ?? 0;\n  if (n < 1) {\n    throw new InvalidParameterError(\n      \"friedmanchisquare() requires all samples to be non-empty\",\n      \"n\",\n      n\n    );\n  }\n\n  for (let i = 1; i < k; i++) {\n    const sample = samples[i];\n    if (sample && sample.size !== n) {\n      throw new InvalidParameterError(\n        \"All samples must have the same length for Friedman test\",\n        \"size\",\n        { expected: n, got: sample.size, sampleIndex: i }\n      );\n    }\n  }\n\n  // Rank within each block (row)\n  const denseSamples = samples.map((sample) =>\n    sample ? toDenseArray1D(sample) : new Float64Array(0)\n  );\n  const rankSums = new Float64Array(k);\n  let tieSum = 0;\n\n  for (let i = 0; i < n; i++) {\n    const block = new Float64Array(k);\n    for (let j = 0; j < k; j++) {\n      const arr = denseSamples[j];\n      block[j] = arr?.[i] ?? 0;\n    }\n    const ranked = rankData(block);\n    tieSum += ranked.tieSum;\n    for (let j = 0; j < k; j++) {\n      const rank = ranked.ranks[j] ?? 0;\n      rankSums[j] = (rankSums[j] ?? 0) + rank;\n    }\n  }\n\n  let chiSq = 0;\n  for (let j = 0; j < k; j++) {\n    const rs = rankSums[j] ?? 0;\n    chiSq += rs * rs;\n  }\n  chiSq = (12 / (n * k * (k + 1))) * chiSq - 3 * n * (k + 1);\n\n  // Tie correction factor for Friedman chi-square.\n  const tieCorrection = n > 0 ? 1 - tieSum / (n * k * (k * k - 1)) : 1;\n  if (tieCorrection <= 0) {\n    throw new InvalidParameterError(\n      \"friedmanchisquare() is undefined when all numbers are identical within blocks\",\n      \"tieCorrection\",\n      tieCorrection\n    );\n  }\n  chiSq /= tieCorrection;\n\n  const df = k - 1;\n  const pvalue = 1 - chiSquareCdf(chiSq, df);\n\n  return { statistic: chiSq, pvalue };\n}\n\n/**\n * Levene's test for equality of variances.\n *\n * Tests whether two or more groups have equal variances.\n * More robust than Bartlett's test for non-normal data.\n *\n * @param center - Method to use for centering: 'median' (default, most robust),\n *                 'mean' (traditional), or 'trimmed' (10% trimmed mean)\n * @param samples - Two or more sample tensors to compare\n * @returns Test result with statistic and p-value\n *\n * @example\n * ```ts\n * import { levene, tensor } from 'deepbox';\n *\n * const group1 = tensor([1, 2, 3, 4, 5]);\n * const group2 = tensor([2, 4, 6, 8, 10]);\n * const result = levene('median', group1, group2);\n * console.log(result.pvalue);  // p-value for equal variances\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.levene.html | SciPy stats.levene}\n */\nexport function levene(center: \"mean\" | \"median\" | \"trimmed\", ...samples: Tensor[]): TestResult {\n  const k = samples.length;\n  if (k < 2) {\n    throw new InvalidParameterError(\"levene() requires at least 2 groups\", \"k\", k);\n  }\n\n  // Convert samples to arrays and compute centers\n  const groups: Float64Array[] = [];\n  const centers: number[] = [];\n\n  for (let g = 0; g < k; g++) {\n    const sample = samples[g];\n    if (!sample || sample.size === 0) {\n      throw new InvalidParameterError(\"levene() requires all groups to be non-empty\", \"groupSize\", {\n        group: g,\n        size: sample?.size ?? 0,\n      });\n    }\n    const arr = toDenseSortedArray1D(sample);\n    if (arr.length < 2) {\n      throw new InvalidParameterError(\n        \"levene() requires at least 2 samples per group\",\n        \"groupSize\",\n        arr.length\n      );\n    }\n    groups.push(arr);\n\n    // Compute center based on method\n    if (center === \"mean\") {\n      let sum = 0;\n      for (let i = 0; i < arr.length; i++) sum += arr[i] ?? 0;\n      centers.push(sum / arr.length);\n    } else if (center === \"median\") {\n      const mid = Math.floor(arr.length / 2);\n      if (arr.length % 2 === 0) {\n        centers.push(((arr[mid - 1] ?? 0) + (arr[mid] ?? 0)) / 2);\n      } else {\n        centers.push(arr[mid] ?? 0);\n      }\n    } else {\n      // Trimmed mean (10%)\n      const trimCount = Math.floor(arr.length * 0.1);\n      let sum = 0;\n      const n = arr.length - 2 * trimCount;\n      for (let i = trimCount; i < arr.length - trimCount; i++) {\n        sum += arr[i] ?? 0;\n      }\n      centers.push(sum / n);\n    }\n  }\n\n  // Compute absolute deviations from center (Z_ij = |Y_ij - center_i|)\n  const Z: Float64Array[] = [];\n  const groupMeansZ: number[] = [];\n  let N = 0;\n  let grandSumZ = 0;\n\n  for (let g = 0; g < groups.length; g++) {\n    const arr = groups[g];\n    if (!arr) continue;\n    const c = centers[g] ?? 0;\n    const zArr = new Float64Array(arr.length);\n    let sumZ = 0;\n\n    for (let i = 0; i < arr.length; i++) {\n      const absVal = Math.abs((arr[i] ?? 0) - c);\n      zArr[i] = absVal;\n      sumZ += absVal;\n    }\n\n    Z.push(zArr);\n    groupMeansZ.push(sumZ / arr.length);\n    N += arr.length;\n    grandSumZ += sumZ;\n  }\n\n  const grandMeanZ = grandSumZ / N;\n\n  // Compute Levene's W statistic (F-test on absolute deviations)\n  let SSB = 0; // Between-group sum of squares\n  let SSW = 0; // Within-group sum of squares\n\n  for (let g = 0; g < Z.length; g++) {\n    const zArr = Z[g];\n    if (!zArr) continue;\n    const n = zArr.length;\n    SSB += n * ((groupMeansZ[g] ?? 0) - grandMeanZ) ** 2;\n\n    for (let i = 0; i < n; i++) {\n      SSW += ((zArr[i] ?? 0) - (groupMeansZ[g] ?? 0)) ** 2;\n    }\n  }\n\n  const dfB = k - 1;\n  const dfW = N - k;\n  if (dfW <= 0) {\n    throw new InvalidParameterError(\n      \"levene() requires more total observations than groups\",\n      \"dfW\",\n      dfW\n    );\n  }\n  if (SSW === 0) {\n    return { statistic: Infinity, pvalue: 0 };\n  }\n  const W = SSB / dfB / (SSW / dfW);\n\n  const pvalue = 1 - fCdf(W, dfB, dfW);\n\n  return { statistic: W, pvalue };\n}\n\n/**\n * Bartlett's test for equality of variances.\n *\n * Tests whether two or more groups have equal variances.\n * Assumes data is normally distributed; use Levene's test for non-normal data.\n *\n * @param samples - Two or more sample tensors to compare\n * @returns Test result with statistic and p-value\n *\n * @example\n * ```ts\n * import { bartlett, tensor } from 'deepbox';\n *\n * const group1 = tensor([1, 2, 3, 4, 5]);\n * const group2 = tensor([2, 4, 6, 8, 10]);\n * const result = bartlett(group1, group2);\n * console.log(result.pvalue);  // p-value for equal variances\n * ```\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.bartlett.html | SciPy stats.bartlett}\n */\nexport function bartlett(...samples: Tensor[]): TestResult {\n  const k = samples.length;\n  if (k < 2) {\n    throw new InvalidParameterError(\"bartlett() requires at least 2 groups\", \"k\", k);\n  }\n\n  // Convert samples to arrays and compute variances\n  const variances: number[] = [];\n  const sizes: number[] = [];\n  let N = 0;\n\n  for (let g = 0; g < k; g++) {\n    const sample = samples[g];\n    if (!sample || sample.size === 0) {\n      throw new InvalidParameterError(\n        \"bartlett() requires all groups to be non-empty\",\n        \"groupSize\",\n        { group: g, size: sample?.size ?? 0 }\n      );\n    }\n    const arr = toDenseSortedArray1D(sample);\n    const n = arr.length;\n    if (n < 2) {\n      throw new InvalidParameterError(\n        \"bartlett() requires at least 2 samples per group\",\n        \"groupSize\",\n        n\n      );\n    }\n\n    // Compute mean\n    let mean = 0;\n    for (let i = 0; i < n; i++) mean += arr[i] ?? 0;\n    mean /= n;\n\n    // Compute sample variance (using n-1 denominator)\n    let ss = 0;\n    for (let i = 0; i < n; i++) {\n      const d = (arr[i] ?? 0) - mean;\n      ss += d * d;\n    }\n    const variance = ss / (n - 1);\n\n    variances.push(variance);\n    sizes.push(n);\n    N += n;\n  }\n\n  // Check for zero variances\n  for (let g = 0; g < k; g++) {\n    if ((variances[g] ?? 0) === 0) {\n      throw new InvalidParameterError(\n        \"bartlett() is undefined when a group has zero variance\",\n        \"variance\",\n        variances[g]\n      );\n    }\n  }\n\n  // Compute pooled variance\n  let pooledNumerator = 0;\n  for (let g = 0; g < k; g++) {\n    pooledNumerator += ((sizes[g] ?? 1) - 1) * (variances[g] ?? 1);\n  }\n  const pooledVariance = pooledNumerator / (N - k);\n\n  // Compute Bartlett's statistic\n  // T = (N-k) * ln(s_p^2) - sum((n_i - 1) * ln(s_i^2))\n  let sumLogVar = 0;\n  for (let g = 0; g < k; g++) {\n    sumLogVar += ((sizes[g] ?? 1) - 1) * Math.log(variances[g] ?? 1);\n  }\n  const T = (N - k) * Math.log(pooledVariance) - sumLogVar;\n\n  // Correction factor C\n  let sumInvDf = 0;\n  for (let g = 0; g < k; g++) {\n    sumInvDf += 1 / ((sizes[g] ?? 1) - 1);\n  }\n  const C = 1 + (1 / (3 * (k - 1))) * (sumInvDf - 1 / (N - k));\n\n  // Chi-square statistic\n  const chiSq = T / C;\n\n  const df = k - 1;\n  const pvalue = 1 - chiSquareCdf(chiSq, df);\n\n  return { statistic: chiSq, pvalue };\n}\n\n/**\n * One-way ANOVA.\n *\n * @see {@link https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.f_oneway.html | SciPy stats.f_oneway}\n */\nexport function f_oneway(...samples: Tensor[]): TestResult {\n  const k = samples.length;\n  if (k < 2) {\n    throw new InvalidParameterError(\"f_oneway() requires at least 2 groups\", \"groups\", k);\n  }\n\n  for (let g = 0; g < k; g++) {\n    const sample = samples[g];\n    if (!sample || sample.size === 0) {\n      throw new InvalidParameterError(\n        \"f_oneway() requires all groups to be non-empty\",\n        \"groupSize\",\n        { group: g, size: sample?.size ?? 0 }\n      );\n    }\n  }\n\n  let N = 0;\n  const means: number[] = [];\n  const sizes: number[] = [];\n  const groups: Float64Array[] = [];\n\n  // Compute group means\n  for (let g = 0; g < k; g++) {\n    const sample = samples[g];\n    if (!sample) {\n      throw new InvalidParameterError(\n        \"f_oneway() requires all groups to be non-empty\",\n        \"groupSize\",\n        { group: g, size: 0 }\n      );\n    }\n    const arr = toDenseArray1D(sample);\n    const n = arr.length;\n    groups.push(arr);\n    sizes.push(n);\n    N += n;\n\n    let sum = 0;\n    for (let i = 0; i < n; i++) {\n      sum += arr[i] ?? 0;\n    }\n    means.push(sum / n);\n  }\n\n  // Compute grand mean\n  let grandSum = 0;\n  for (let g = 0; g < k; g++) {\n    grandSum += (means[g] ?? 0) * (sizes[g] ?? 0);\n  }\n  const grandMean = grandSum / N;\n\n  // Compute between-group and within-group variance\n  let SSB = 0; // Between-group sum of squares\n  let SSW = 0; // Within-group sum of squares\n\n  for (let g = 0; g < groups.length; g++) {\n    const arr = groups[g];\n    if (!arr) continue;\n    const n = arr.length;\n    SSB += n * ((means[g] ?? 0) - grandMean) ** 2;\n\n    for (let i = 0; i < n; i++) {\n      SSW += ((arr[i] ?? 0) - (means[g] ?? 0)) ** 2;\n    }\n  }\n\n  const dfB = k - 1;\n  const dfW = N - k;\n  if (dfW <= 0) {\n    throw new InvalidParameterError(\n      \"f_oneway() requires at least one group with more than one sample\",\n      \"dfW\",\n      dfW\n    );\n  }\n  const MSB = SSB / dfB;\n  const MSW = SSW / dfW;\n  if (MSW === 0) {\n    // All within-group values are identical; F is infinite if groups differ, NaN otherwise\n    const F = MSB === 0 ? NaN : Infinity;\n    return { statistic: F, pvalue: MSB === 0 ? NaN : 0 };\n  }\n  const F = MSB / MSW;\n\n  const pvalue = 1 - fCdf(F, dfB, dfW);\n\n  return { statistic: F, pvalue };\n}\n"]}